Clinical application of 2D perfusion angiography in critical ...

ADVERTIMENT. Lʼaccés als continguts dʼaquesta tesi queda condicionat a lʼacceptació de les condicions dʼúsestablertes per la següent llicència Creative Commons: http://cat.creativecommons.org/?page_id=184

ADVERTENCIA. El acceso a los contenidos de esta tesis queda condicionado a la aceptación de las condiciones de usoestablecidas por la siguiente licencia Creative Commons: http://es.creativecommons.org/blog/licencias/

WARNING. The access to the contents of this doctoral thesis it is limited to the acceptance of the use conditions setby the following Creative Commons license: https://creativecommons.org/licenses/?lang=en

CLINICAL APPLICATION OF 2D PERFUSION ANGIOGRAPHY

IN CRITICAL LIMB ISCHEMIA REVASCULARIZATION

EFREM GÓMEZ JABALERA Universitat Autònoma de Barcelona

2020

2

Cover illustration by August Coromines, MD, for “Peuàs”. Barcelona, Catalunya, Spain (2015).

3

Clinical application of 2D

perfusion angiography in

critical limb ischemia

revascularization Aplicació clínica de l’angiografia de perfusió 2D en el tractament de la isquèmia crítica de l’extremitat

Aplicación clínica de la angiografía de perfusión 2D en el tratamiento de la isquemia crítica de la extremidad

Efrem Gómez Jabalera

DOCTORAL THESIS – UAB 2020

Thesis directors: Dr. Artigas Raventós, Vicenç Dr. Bellmunt Montoya, Sergi Dr. Dilmé Muñoz, Jaume Thesis tutor: Dr. Artigas Raventós, Vicenç

Doctorat en Cirurgia i Ciències Morfològiques Facultat de Medicina

Departament de Cirurgia Universitat Autònoma de Barcelona

5

A mi abuelo

7

Dr. Vicenç Artigas Raventós, MD, PhD, Surgical Department, Univesitat Autònoma de Barcelona,

Barcelona; Dr. Sergi Bellmunt Montoya, MD, FEBVS, PhD, Hospital Universitari Vall d'Hebron,

Univesitat Autònoma de Barcelona and Dr. Jaume Dilmé Muñoz, MD, FEBVS, PhD, Hospital de

la Santa Creu i Sant Pau, Univesitat Autònoma de Barcelona, Spain.

They declare that Efrem Gómez Jabalera has performed under their supervision, the studies

presented in the Dissertation Clinical application of 2D perfusion angiography in critical limb

ischemia revascularization. This Dissertation has been structured following the normative for PhD

Thesis to obtain the degree of International Doctor Mention and it is ready to be presented to a

Tribunal.

Barcelona, June 2020

Director and Tutor Director Director

Vicenç Artigas Raventós Sergi Bellmunt Montoya Jaume Dilmé Muñoz

9

No sabía

si era un limón amarillo

lo que tu mano tenía,

o el hilo de un claro día,

Guiomar, en dorado ovillo.

Tu boca me sonreía.

Yo pregunté: ¿Qué me ofreces?

¿Tiempo en fruto, que tu mano

eligió entre madureces

de tu huerta?

¿Tiempo vano

de una bella tarde yerta?

¿Dorada esencia encantada?

¿Copla en el agua dormida?

¿De monte en monte encendida,

la alborada

verdadera?

¿Rompe en sus turbios espejos

amor la devanadera

de sus crepúsculos viejos?

Antonio Machado

11

AGRADECIMIENTOS / ACKNOWLEDGMENTS

Agradezco ante todo a todas las personas que me han apoyado y confiado durante todo

este proyecto, y en especial a mi amada Guiomar, que, con su ejemplo, comprensión y no

pocos esfuerzos dedicados a mi, es la compañera de vida ideal. También doy las más sentidas

gracias a toda mi familia: Mamá, Papá, Víctor, Amets, Martina, Marta, Anders, Edgard, Mia,

Lea, Irene y Romà; ya que no soy sino parte de todos ellos y ellos parte de mi. También

agradecer a Lina, Carlos y Juan Carlos por haberme inspirado en la actitud del esfuerzo y la

excelencia. Y no podrían faltar mis infinitos agradecimientos a todos mis incondicionales

amigos y amigas; con quien he crecido y son un pilar esencial.

Tampoco hubiera conseguido nada sin el servicio de Angiología y Cirugía Vascular

del Hospital de la Santa Creu i Sant Pau, quienes me introdujeron e inspiraron en la Cirugía

Vascular y Endovascular, razón por la que Profesor Dr. Bellmunt y Profesor Dr. Dilmé, dos

de mis principales maestros también les debo el agradecimiento de haberme apoyado como

directores de este proyecto. Del mismo modo agradezco, del Departamento de Cirugía de la

Universitat Autònoma de Barcelona, al Profesor Dr. Artigas, también tutor, apasionado por la

docencia y el avance de la Medicina y la Cirugía. También debo especial mención a mis

maestros y referentes Dr. August Corominas, Dr. Gaspar Mestres y Dr. Jordi Villalba; sin

obviar el apoyo que he recibido de Dr. Josep María Mestres y el equipo de Angiogrup; y a Dr.

Vicenç Riambau y Dr. Joan Martínez Benazet por su confianza en estos primeros años como

especialista.

Obviamente, el proyecto no hubiera podido ni siquiera ser planteado sin el trabajo y

valor profesional de Dr. Mariano Palena y Dr. Marco Manzi, con todo su equipo de la Casa

di Cura Policlínico Abano Terme, por haberme acogido y enseñado tantísimo, además de

acogerme desinteresadamente como su fellow, una parte imprescindible del proyecto. Por

último, quiero agradecer al Dr. Vicente Medina su apoyo en el planteamiento del trabajo y

ayudarme a ampliar mi rango de visión hacia las posibilidades del “por qué” de este proyecto.

A todo el mundo que menciono explícita e implícitamente, les doy las más absolutas

gracias y mi compromiso que este proyecto es sólo el punto de partida de tantas otras

búsquedas de respuestas y mejorías en la atención a los pacientes.

13

TABLE OF CONTENTS

15

Table of contents _____

I. Summary / Resum / Resumen…………………………………………………………….. 19

II. Abbreviations ……………………………………………………………………………. 27

III. Conflicts of interest and financial supports……………………………………………… 31

1. INTRODUCTION ………………………………………………………………… 35

1.1. Peripheral Arterial Disease. …………………………………………………… 37

1.1.1. Definition and epidemiology of PAD……………………………. 37

1.1.2. Critical limb ischemia …………………………………………… 38

1.1.3. Risk factors for PAD. …………………………………………… 39

1.1.4. Pathophysiology of PAD ………………………………………… 42

1.2. Present classification and risk stratification systems of PAD. ………………… 48

1.2.1. Leriche-Fontaine classification. …………………………………. 48

1.2.2. Rutherford classification.………………………………………… 48

1.2.3. University of Texas ulcer classification. ………………………… 50

1.2.4. TASC anatomical classifications. ...……………………………… 51

1.2.5. SVS Wound Ischemia foot Infection (WIfI) stages ……………… 52

1.3. Present methods for ischemic assessment…………………………………….… 55

1.3.1. Ankle Brachial Index and ankle pressure.………………………… 56

1.3.2. Toe pressure ………………………………………………………. 56

1.3.3. Pulse volume recordings. ………………………………………… 57

1.3.4. Photoplethysmography …………………………………………… 57

1.3.5. Transcutaneous oxygen pressure (TcPO2) ………………………… 58

1.3.6. Tissue oxygen saturation mapping………………………………… 59

1.3.7. Hyperspectral imaging. …………………………………………… 59

1.3.8. Oxygen-to-see (O2C) method……………………………………… 61

1.3.9. Micro Oxygen sensors (MOXYs) ………………………………… 62

1.3.10. Indocyanine green fluorescence imaging…………………………… 62

1.3.11. Peripheral duplex ultrasound imaging……………………………… 64

1.3.12. Tomographic ultrasound imaging systems………………………… 65

1.3.13. Computerized tomography angiography…………………………… 65

1.3.14. Magnetic resonance imaging angiography………………………… 66

1.4. Natural history of critical limb ischemia. ……………………………………… 66

1.4.1. Treatment and management of critical limb ischemia ………….… 67

1.4.2. Endpoints of treatment. …………………………………………… 71

1.5. 2D Perfusion angiography.……………………………………………………… 73

2. JUSTIFICATION…………………………………………………………………… 77

16

3. HYPOTHESIS AND OBJECTIVES. ………………………………………………. 81

4. METHODS. ………………………………………………………………………… 85

4.1. Study design. …………………………………………………………………… 87

4.1.1. Setting …………………………………………………………… 87

4.1.2. Patient selection…………………………………………………… 87

4.1.3. Inclusion criteria…………………………………………………… 87

4.1.4. Exclusion criteria. …………………………………………………. 87

4.1.5. Interventions ………………………………………………………. 88

4.1.6. Variables …………………………………………………………… 88

4.1.7. Clinical follow-up ………………………………………………… 89

4.1.8. Data collection ……………………………………………………. 90

4.1.9. Ethics ……………………………………………………………… 90

4.2. Perfusion angiography measurement…………………………………………… 90

4.2.1. Projection…………………………………………………………. 90

4.2.2. ROI tracing. ……………………………………………………… 91

4.2.3. Time of exposure ………………………………………………… 91

4.2.4. Artifact management……………………………………………… 92

4.3. Conventional angiography: measurement and classification…………………… 93

4.3.1. TASC classification. ……………………………………………… 93

4.3.2. Hemodynamic angiography based “Abano Terme Score” ………… 94

4.4. Endpoints and groups for analysis ………………………………………………. 96

4.5. Statistical analysis ………………………………………………………………. 97

5. RESULTS …………………………………………………………………………… 99

5.1. Description of the population …………………………………………………… 101

5.2. Distribution of the endpoints and PA parameters………………………………… 104

5.3. Follow-up…………………………………………………………………………. 108

5.4. Predictive ability of clinical success in patients with rest pain…………………… 109

5.5. Predictive ability of clinical success in patients with ulcers……………………… 110

5.5.1. Multivariate analysis and confounding factors……………………… 114

5.6. Correlations of PA parameters with TASC and ATS……………………………… 115

5.7. Non-invasive diagnostic methods………………………………………………… 116

5.8. Effects of incomplete plantar arch on PA parameters ……………………………. 117

6. DISCUSSION………………………………………………………………………… 119

6.1. Results compared with other perfusion angiography studies ……………………. 121

6.2. Internal and external validity ..…………………………………………………. 124

6.3. Interpretation of the results, possible confounding factors and clinical translation ………………………………………………………………… 125

6.4. Comparison with other diagnostic methods……………………………………… 127

17

6.5. Secondary results………………………………………………………………… 131

6.5.1. Disease burden anatomical classification. ………………………… 131

6.5.2. TcPO2 correlation with perfusion angiography…………………… 133

6.5.3. Effects of infrapopliteal anatomical variations …………………… 133

6.5.4. Effects of a cropped plantar arch ………………………………… 134

6.6. Limitations and further possibilities ……………………………………… 134

7. CONCLUSIONS …………………………………………………………………… 137

8. REFERENCES……………………………………………………………………… 141

9. List of figures………………………………………………………………………… 157

10. List of tables ………………………………………………………………………… 161

11. ANNEXES. ………………………………………………………………………… 165

11.1. Examples of perfusion angiography images. ………………………………. 167

11.2. Presented abstract: Hemodynamic significance of perfusion angiography parameters ………………………………………………………………………. 170

11.3. Presented abstract: Perfusion angiography in the prediction of wound healing in endovascular treatment of critical limb ischemia ……………………………….. 171

11.4. Presented abstract: Angiography based Abano Terme Score: a novel descriptive system for below the groin arterial disease.……………………………………………… 172

11.5. Article submitted to the European Journal of Vascular & Endovascular Surgery: Novel Abano Terme Scores and its application in the treatment of critical limb threatening ischemia. …………………………………………………………………………...…… 173

19

I. SUMMARY

RESUM

RESUMEN

21

Summary

The endovascular treatment (EVT) of Peripheral Arterial Disease (PAD) is based on

angiographic imaging and post-revascularization treatment success is based on the subjective

interpretation of this visual assessment. 2D perfusion angiography (PA) is an image-

processing software which may allow for the quantification of perfusion. In addition, a

simpler anatomic classification system, able to describe the arterial disease burden below the

groin, needs to be designed to determine the best therapy for any given patient.(1)

The aim of this thesis is to create an objective system to assess the success of EVT

based on the quantification of tissue perfusion through PA, capable of accurately predicting

the healing probability of ischemic ulcers. Secondly, we seek to describe a classification

system of easy application during daily clinical practice that will also facilitate comparison of

patients among clinical trials.

The Project was designed as a retrospective cohort study with consecutive patients

undergoing EVT at a single specialized center for critical limb ischemia (CLI). The cases were

analyzed with PA before and after treatment, and also ranked according to current

classification systems (Rutherford, TASC and WIfI) and a new proposed classification: the

Abano Terme Score (ATS). Demographic and clinical data were recorded and clinical follow-

up was performed (at 1 and 6 months). The PA parameters were Arrival Time (AT), Peak

Time (PT), Wash-in Rate, Width, Area Under Curve and Mean Transit Time (MTT). Two

cohorts were defined based upon a time to heal of less or longer than 30 days.

From January 2015 to July 2016, PA analysis was performed on 580 consecutive

patients that underwent EVT. Among them, 332 met the inclusion criteria to be studied, from

which 123 were excluded for ulcer healing analysis (34 because of poor image quality, 50

patients had no ulcer, 20 died and 19 were lost at follow-up). Mean age was 72 years and 67.5%

were men; 133 patients had Rutherford 5 and 76 had Rutherford 6 lesions, with similar

distribution in both groups. The WIfI risk for amputation was also similar for both groups,

and it was low in 24%, moderate in 14% and high in 62%. We found significant differences

between groups in the healing time for the following cut-off values of PA parameters: AT>6

seconds and improvement of MTT>1.7 seconds or the MTT>4.1 seconds after the treatment.

The ATS, while being a simpler classification than current used system, not only showed a

better correlation with parameters such as the transcutaneous pressure of oxygen (TcPO2) and

PA; but also demonstrated, in a subsequent analysis, a better correlation with ulcer healing

and amputation free-survival in patients with Rutherford 5 lesions.

22

Resum

El tractament endovascular de la malaltia arterial perifèrica està basat en les imatges

angiogràfiques. La interpretació subjectiva feta a partir d’elles és com més habitualment

s’avalua l’èxit del tractament durant la revascularització. L’angiografia de perfusió 2D (AP)

és un software de processat de la imatge que podria donar peu a la quantificació de la perfusió

distal. Endemés, un sistema de classificació anatòmic més senzill, capaç de descriure la

càrrega de malaltia arterial per sota de l’engonal, és necessari per poder triar el millor

tractament per un pacient concret.

La intenció d’aquesta tesis és identificar una mesura objectiva per avaluar l’èxit del

tractament endovascular basant-se en la quantificació de la perfusió del teixit amb l’AP; i per

consegüent tenir la capacitat de predir amb major precisió la curació de les úlceres

isquèmiques. Secundàriament, hem buscat adaptar un sistema de classificació que pugui ser

aplicat amb facilitat a la pràctica clínica diària i que permeti la comparació entre pacients a

assajos i estudis clínics.

La investigació del projecte present es va realitzar amb un estudi de cohorts

retrospectiu amb pacients consecutius sotmesos a tractament endovascular en un únic centre

especialitzat en tractament de la isquèmia crítica de l’extremitat. Als pacients se’ls va realitzar

una AP abans i després del tractament, i també es van classificar d’acord amb els sistemes de

classificació més utilitzats (Rutherford, TASC i WIfI) i amb una nova classificació proposta:

l’score d’Abano Terme (ATS). Les dades demogràfiques i clíniques es van recollir i es va

realitzar un seguiment clínic al mes i als 6 mesos. Els paràmetres de l’AP van ser el temps

d’arribada (AT), el temps de pic (PT), la velocitat del rentat (WS), l’amplada (W), l’àrea sota

la corba (AUC) i el temps de trànsit mig (MTT). Les dos cohorts es van definir en base a un

temps de curació major o menor a 30 dies.

Des del gener de 2015 fins al juliol de 2016, 580 pacients consecutius van ser tractats

endovascularment i van ser estudiats amb AP. D’entre ells, 332 complien els criteris

d’inclusió, dels quals 123 van ser exclosos posteriorment de l’anàlisi per la curació de les

úlceres (en 34 casos la imatge de l’AP tenia mala qualitat, 50 pacients no presentaven úlceres,

20 van finar i 19 no van completar el seguiment). L’edat mitja va ser de 72 anys i el 67,5%

eren homes. 133 pacients presentaven lesions Rutherford 5 i en 76 les lesions eren Rutherford

6. El risc WIfI d’amputació va ser baix en un 24%, moderat en un 14% i alt en un 62%. Vam

trobar taxa de curació als 30 dies, amb diferències estadísticament significatives entre els

grups, per als següents punts de tall dels paràmetres de l’AP: AT > 6 segons i MTT > 4.1

23

segons o un increment del mateix MTT > 1,7 segons. L’ATS, sent un sistema de classificació

més senzill que es existents actualment, no només es va correlacionar millor que amb la

TcPO2 i amb l’AP; sinó que en un posterior anàlisis va demostrar millor correlació amb la

curació de les úlceres i la supervivència lliure d’amputació en pacients amb lesions Rutherford

5.

24

Resumen (Castellano)

El tratamiento endovascular de la enfermedad arterial periférica está basado en las

imágenes angiográficas. La interpretación subjetiva hecha a partir de ellas es el modo más

habitual de evaluar el éxito del tratamiento durante la revascularización. La angiografía de

perfusión 2D (AP) es un software de procesado de la imagen que podría permitir la

cuantificación de la perfusión distal. Además, un sistema de clasificación anatómico más

sencillo, capaz de describir la carga de enfermedad arterial por debajo de la ingle, es necesario

para decidir el mejor tratamiento para un paciente dado.

La intención de esta tesis es identificar una medida objetiva para evaluar el éxito del

tratamiento endovascular basado en la cuantificación de la perfusión del tejido con la

angiografía de perfusión; y por ende ser capaz de predecir con mayor precisión la curación de

las úlceras isquémicas. Secundariamente, hemos tratado de adaptar un sistema de clasificación

que pueda ser aplicada fácilmente a la práctica clínica diaria y que permita comparaciones

entre pacientes en ensayos clínicos y estudios.

La investigación de este proyecto se basó en un estudio de cohortes retrospectivo con

pacientes consecutivos sometidos a tratamiento endovascular en un único centro

especializado para el tratamiento de la isquemia crítica de la extremidad. Los pacientes fueron

analizados con AP antes y después del tratamiento, y también se clasificaron de acuerdo con

los sistemas más utilizados (Rutherford, TASC y WIfI) y con una nueva clasificación

propuesta: el score de Abano Terme (ATS). Los datos demográficos y clínicos se recogieron

y se realizó un seguimiento clínico a al primer mes y a los 6 meses. Los parámetros de la AP

fueron tiempo de llegada (AT), el tiempo de pico (PT), la velocidad de lavado (WS), la

amplitud (W), el área bajo la curva (AUC) y el tiempo de tránsito medio (MTT). Las dos

cohortes se definieron en base a un tiempo de curación de menor o mayor a 30 días.

De enero de 2015 a julio de 2016, 580 pacientes consecutivos se sometieron a un

tratamiento endovascular, realizándose en ellos un análisis con AP. Entre ellos, 332

cumplieron los criterios de inclusión, de los cuales 123 se excluyeron del análisis de curación

de úlceras (34 debido a la mala calidad de la imagen de AP, 50 pacientes no presentaban

úlceras, 20 fallecieron y 19 no completaron el seguimiento). La edad media fue de 72 años y

el 67,5% eran hombres. 133 pacientes presentaban lesiones Rutherford 5 y en 76 las lesiones

eran Rutherford 6. El riesgo WIfI de amputación fue bajo en 24%, moderado en 14% y alto

en 62%. Encontramos una tasa curación a los 30 días , con diferencias estadísticamente

significativas entre los grupos, para los siguientes valores de corte de los parámetros de la AP:

25

AT > 6 segundos y MTT> 4.1 segundos o un incremento del mismo MTT > 1.7 segundos. El

ATS, siendo un sistema de clasificación más simple que los actualmente empleados, no sólo

se correlacionó mejor con la TcPO2 y la angiografía de perfusión; sino que en un posterior

análisis demostró una mejor correlación con la curación de las úlceras y la supervivencia libre

de amputación en pacientes con lesiones Rutherford 5.

27

II. ABBREVIATIONS

29

Ankle Brachial Index ABI

Atrial Fibrillation AF

Amputation Free Survival AFS

Arrival Time AT

Abano Terme Score ATS

Area Under the Curve AUC

Body Mass Index BMI

Below-The-Knee BTK

Coronary Artery Bypass Graft CABG

Coronary Artery Disease CAD

Cardiovascular Disease CD

Common Femoral Artery CFA

Chronic Kidney Disease CKD

Critical Limb Ischemia CLI

Cerebro-Vascular Disease CVD

Diabetes Mellitus DM

Diabetic Foot Ulcer DFU

Digital Subtraction Angiography DSA

Duplex ultrasound DUS

End-Stage Renal Disease ESRD

Endovascular Treatment EVT

Femoro-popliteal FP

Global Limb Anatomic Staging System GLASS

Great Saphenous Vein GSV

Hyperspectral imaging HI

High-Density Lipoprotein Cholesterol HDL-C

Indocyanine Green Fluorescence Imaging ICG-FI

Ischemic CardioMyopathy ICM

Joint Vascular Societies Council JVSC

Laser Doppler LD

Medial Arterial Calcification MAC

Major Adverse Cardiovascular Event MACE

Major Adverse Limb Event MALE

Multispectral optoacoustic tomography MSOT

Micro OXYgen sensor MOXY

Mean Transit Time MTT

Objective Performance Goals OPG

Perfusion Angiography PA

Peripheral Arterial Disease PAD

Percutaneous Coronary Intervention PCI

Peak Time PT

Relative Hemoglobin rHb

Region Of Interest ROI

Saturation of oxygen StO2

Secondary HyperParaThyroidism SHPT

Superficial Femoral Artery SFA

TransCutaneous Pressure of Oxygen TcPO2

Target Lesion Revascularization TLR

Time to Peak TP

TibioPeroneal Trunk TPT

Time to heal TTH

Texas Ulcer Classification TUC

Toe Pressure TP

Vascular Calcification VC

Vascular Smooth Muscle Cells VSMCs

Width W

Wound Ischemia Foot Infection WIfI

Wash Speed WS

31

III. CONFLICTS OF INTEREST

AND

FINANCIAL SUPPORTS

33

The author reports no financial relationships or conflicts of interest regarding the content

herein. However, Efrem Gómez Jabalera acknowledges a mobility grant from the Government of

Andorra, AMXXX-AND-2017”.

35

1. INTRODUCTION

37

1.1. Peripheral Arterial Disease

1.1.1. Definition and epidemiology of PAD

Peripheral artery disease (PAD) addresses chronic atherosclerotic disease that partially

or completely obstructs ≥ 1 peripheral arteries.(2) It mainly affects the arteries of the lower

limbs, but also the carotid and visceral arteries, and the arteries of the upper limbs.

The mechanism underlying PAD is an insufficient blood supply to the legs that

provokes pain and muscular dysfunction, similar to that in coronary angina. Symptoms occur

when walking and are relieved at rest in the case of intermittent claudication (IC); disease

severity and the patient’s degree of activity may modify clinical presentation.(3) Seventy-five

percent of patients remain clinically stable due to the development of collaterals, metabolic

muscular adaptation and/or the patients themselves altering the gait to favor non-ischemic

muscle groups.

In the more severely affected cases, the pain may occur at rest, associating tissue loss

and/or necrosis. As graded by Rutherford’s and Leriche-Fotaine’s classifications, this

situation is known as critical limb ischemia (CLI): grade 4 in Rutherford’s or grade III in

Leriche-Fontaine’s classification (pain at rest) and grades 5 and 6 in Rutherford’s or grade IV

in Leriche-Fontaine’s classification (which meet the presence of an ischemic ulcer or

gangrene). Critical limb ischemia (CLI) was first defined in published form in 1982,(4) and

the threshold criteria have slightly changed.

It is estimated that >200 million people have PAD worldwide, with a spectrum from

none to severe symptoms.(5) It is uncommon before the age of 50, and its prevalence rises

sharply onwards. It is estimated that up to 23% of patients older than age 55 may suffer from

PAD.(6) Data from Framingham Study show incidence increasing from <0.4 per 1000 in men

35-45 years to 6 per 1000 in men aged > 65 years.(7) Patients with CLI comprise

approximately 2-3% of all cases of PAD, and the diagnosis portends a dismal prognosis.(8)

From 2003 to 2008, the annual incidence of PAD and CLI were 2.35% and 0.35%,

respectively; corresponding to a prevalence of 10.69% for PAD, and 1.33% for CLI.

Disease definition, that has evolved over time, is particularly important to precisely

describe prevalence and incidence rates. Earlier definitions were focused on IC, using Rose

criteria (an standardized questionnaire aimed at identifying IC symptoms); nowadays the

Ankle Brachial Index (ABI) value of ≤ 0.90 is more widely used to define the disease.(3)

Nevertheless, there are few epidemiological studies of PAD based on ABI, given the time and

38

resources required to periodically retest study subjects for incident disease. In the Limburg

peripheral arterial occlusive disease Study, incidence rates for PAD were based on 2 ABI

measurements of <0.95.(9)

Compared to other cardiovascular diseases (CVDs), rates are more similar for men

and women. Among men, annual incidence of PAD was 1.7 per 1000 at the age of 40 to 54

years; 1.5 per 1000 at the age of 55 to 64; and 17.8 per 1000 at the age of ≥65. Annual

incidence in women was higher: 5.9, 9.1, and 22.9 per 1000, respectively, for the same age

groups.(9) In the Framingham Offspring Study, PAD based on ABI was found in 3.9% of

men and 3.3% of women, for a ratio of 1.18.(10) On the other hand, in the Rotterdam Study,

PAD was actually lower in men than in women, with prevalence of 16.9% and 20.5%, with a

ratio of men:women of 0.82.(11) Still another population-based study from Southern Italy

found prevalence of PAD to be similar in men and women, with male to female ratios from

0.89 to 0.99, depending on the age.(12) Finally, the Cardiovascular Health Study, an ABI of

<0.90 was somewhat more prevalent in men (13.8%) than in women (11.4%; ratio, 1.21), but

the association of disease with sex was not significant after adjustment for age and CVD

status.(13)

1.1.2 Critical limb ischemia

The first description of CLI was made in 1982 by Earnshaw et al.(4) and intended to

be applied only to non-diabetic patients. The threshold for CLI was set at an ankle pressure

(AP) <40 mm Hg in the presence of rest pain and <60 mm Hg in the presence of tissue loss.

Later on, at the European Consensus of 1992, CLI was defined as rest pain for more than 2

weeks, ulceration or gangrene, and an AP <50 mmHg or a toe pressure (TP) of <30 mmHg,

including diabetic patients.(14) However, that definition is not universally supported, because

the pressures required of healing in the presence ulceration may be higher. The Trans-Atlantic

Inter-Society Consensus (TASC) II suggests that an ankle pressure of <70 mmHg or a TP <50

mmHg is more accurate for CLI condition.(15) Yet another definition is described by

Vallabhaneni et al.(16), that found that CLI in diabetic patients relies mainly on the TP <30

mmHg, while a TP between 30 and 50 mmHg was associated with a low risk of limb loss if

untreated. Despite all the efforts to attain a precise hemodynamic definition, up to one third

of the patients with clinically determined CLI did not meet the usual hemodynamic

criteria .(17)

39

In order to provide a descriptive and predictive stratification of patients, the Society

for Vascular Surgery described the Wound, Ischemia and foot Infection (WIfI) classification

system. Based on patients’ lower extremity loss risk with respect to the natural history of the

disease (if treated conservatively) and depending on the benefit of different treatment

strategies, the WIfI classification refocuses the approach to the patient with a threatened limb

and a component of chronic ischemia according to clinical disease severity.(1) According to

the WIfI classification each of the three components (wound, ischemia and foot infection) are

classified into grades according to specific criteria. Depending on the 3 components’ grades,

classes for each patient are assigned. The spectrum of symptoms is organized by a Delphi

Consensus process into four clinical stages that grade both the risk of amputation and the

benefit of revascularization, assuming the need for a prospective validation. Several studies

have validated the WIfI classification for critical limb threatening ischemia as a predictor for

limb survival and amputation free survival,(18–22) for time to wound healing(23) and

subsequently for cost of care for diabetic foot ulcers.(24) However, the risk of major

amputation at 1 year could be more dependent on the multidisciplinary approach of diabetic

foot ulcers than the WIfI initial stage and the time for wound healing.(25)

1.1.3. Risk factors for PAD

The main risk factors are diabetes, smoking, chronic kidney disease, aging,

hypertension, and dyslipidemia. In the following paragraphs all of them are exposed.

Diabetes Mellitus

The number of people with diabetes has risen from 108 million (prevalence of 4.7%)

in 1980 to 422 million (prevalence of 8.5%) in 2014, as reported by the WHO.(26) This

growth in prevalence has appeared more rapidly in middle- and low-income countries.

Diabetes is a major cause of blindness, kidney failure, heart attacks, stroke and lower limb

amputation. In 2015 an estimated 1.6 million deaths were attributed to high fasting blood

sugar levels, and almost half of them occurred before the age of 70 years. It is projected that

by 2030, diabetes will be the 7th leading cause of death worldwide.

In patients with diabetes, for every 1% increase in hemoglobin A1c there is a

corresponding 26% increased risk of PAD.(27) More severe or longstanding diabetes mellitus

seems to be more strongly related to PAD,(3) but even insulin resistance without diabetes

raises the risk of PAD approximately 40-50%, as it plays a key role in the clustering of

cardiometabolic risk factors which include hyperglycemia, dyslipidemia, hypertension and

40

obesity. PAD in diabetic patients has an aggressive presentation, with early large vessel

involvement coupled with distal microvascular disease and neuropathy.(15) Regardless of

ischemia, the relative immunosuppression in diabetes increases the risk of amputation upon

infection.(3) Amputation is 5 to 10 times more frequent in diabetics,(15) while death is 3 times

higher.(28)

Smoking

Smoking, which is a modifiable risk factor, is one of the strongest risk factors for PAD.

Indeed, active smoking has been shown to confer 2-4 times greater risk of PAD versus

nonsmoking.(3) In addition, smoking cessation has been shown to improve the progression of

PAD, as well as improving intermittent claudication or reducing mortality.(29) Regarding

passive smoking and PAD, in a study conducted on women from Beijing who had never

smoked, the hazard ratio (HR) for PAD was at 1.67, with a significant dose-response

relationship.(30) To sum up, smoking is associated with a significantly higher relative risk for

PAD compared to other atherosclerotic diseases.(31)

Chronic Kidney Disease

Chronic kidney disease (CKD) defined according to creatinine levels and particularly

in the case of end-stage renal disease (ESRD) requiring dialysis has been associated with PAD

in several studies.(3) Using β2-Microglobulin and cystatin C as markers of renal function,

PAD could be predicted in men but not in women.(32) However, when measuring renal

function with MDRD-4, CKD appeared to be a better predictor of mortality or myocardial

infarction than other risk factors in patients with PAD.(33)

Aging

Supporting the data mentioned in the epidemiology section, the prevalence of PAD

rises rapidly in population over 65 years of age.(15) As defined by an ABI of <0.9, the

prevalence ranged from 2.5% in the age group 50–59 years to 14.5% in subjects > 70 years.(34)

Hypertension

The association of hypertension with PAD has been demonstrated in most studies in

which blood pressure was studied. The odds ratio for hypertension ranged from 1.50 to 2.20,

being the systolic blood pressure independently associated with PAD.(3) In the Framingham

41

Study, 30% of the risk of IC in the population was attributable to blood pressure >160/100

mm Hg.(35)

Dyslipidemia

In many studies, high lipid levels were significantly associated with PAD in

multivariable analysis; in fact, 17% of PAD prevalence is attributable to hypercholesterolemia.

Nevertheless, in some other studies, dyslipidemia was significant in the univariate analysis

but dropped out of multivariable models in which other lipid measures were considered. High-

density lipoprotein cholesterol (HDL-C) has been shown to be protective against PAD in most

studies where it was evaluated.(3)

There is also some evidence suggesting that elevated triglycerides may play a role in

disease progression or more severe PAD, and they could be independently associated with

PAD since triglycerides frequently drop out as an independent risk factor.(3)

In recent studies, total cholesterol to high-density lipoprotein cholesterol (HDL-C)

ratio is considered the best lipid measure of risk.(36) Despite all this, the dyslipidemia profile

observed in insulin resistance and diabetes with low HDL-C and high triglycerides, is also

associated with PAD.(3)

Other risk factors

Obesity as a risk factor fails to support a consistent, independent positive association

with PAD. (3) Even in some studies a protective effect was found for claudication in men and

seemed to have a U-shaped, nonlinear relationship with relative weight in women.(3,37) In

any case, several studies found that central adiposity (by measuring waist/hip ratio rather than

BMI), particularly in diabetic patients, was associated with significantly higher risk of PAD,

as in coronary artery disease (CAD).(38,39)

Light-to-moderate alcohol consumption, as observed in CAD, is less consistent with

PAD. 5 Due to the frequent association of alcohol consumption with other risk factors, the

measure of its effects is uncertain. Depending on the population group it could differ. In

Native Americans, a protective effect of alcohol was seen even in multivariable analysis,(40)

but in elderly Japanese American men, alcohol intake was found to increase the risk of

incident PAD.(41)

42

Several studies suggest a higher risk of PAD among Blacks and American Indians

versus Non-Hispanic Whites, whereas there is a lower prevalence among Asians and

Hispanics.(3)

The importance of homocysteine as a risk factor for PAD has been examined in many

studies, with conflicting results. Other inflammatory markers, such as the C-reactive protein

(CRP) and fibrinogen, have been shown, on the other hand, to be associated with PAD in

many studies.(3) CRP, the neutrophil-to-lymphocyte ratio, homocysteine, and the urinary

albumin-to-creatinine ratio appeared to be associated with higher mortality rates.(42) The

neutrophil-to-lymphocyte ratio is found to be able to predict PAD and an elevated platelet-

to-lymphocyte ratio can predict osteomyelitis in diabetic foot ulcers.(43)

1.1.4. Pathophysiology of PAD

Arterial wall composition

The arterial vascular wall is a dynamic tissue that is able to adapt and reorganize itself

under both physiologic and pathologic stimuli. All arterial vessels except capillaries are

composed of three concentric layers with distinct cell and interstitial composition (figure

1):(44)

§ Intima, the innermost layer, is in contact with the blood flow. The intima layer is

composed by a monolayer of endothelial cells, a very thin basal lamina and a

subendothelial layer formed by collagen and elastic fibrils.

§ Media, the middle layer of the vascular wall, is composed by vascular smooth muscle

cells, collagen and a network of elastic fibrils. It is separated from the intima and the

adventitia layers by the internal and external elastic lamina, respectively.

§ Adventitia, the external layer of the vascular wall, consists of elastic fibers, fibroblast,

collagen, nerves, and small blood vessels called vasa vasorum.

43

Figure 1. Arterial wall structure. The arterial wall consists of three concentric layers, called tunica intima,

media and adventitia, separated by elastic membranes. (Courtesy of Prof. Badimon).

Atherosclerosis

Atherosclerosis is a chronic inflammatory disease of the vascular wall, produced by

lipid infiltration, foam macrophage accumulation on the inner wall, and subsequent focal

thickening of the intimal layer. Lesion rupture leads to local thrombus formation, arterial

occlusion, and tissue ischemia and necrosis, with disfunction of the affected organ. The

atherosclerotic and thrombotic processes, with their clinical complications, appear

interdependent and, therefore, are integrated under the term atherothrombosis (figure 2).(45)

Figure 2. Atherothrombotic disease progression. Atherothrombosis chronically develop from early fatty

streak to atheromatous plaque formation that by unpredictable disruption leads to platelet activation and

thrombus formation. (Courtesy of Prof. Badimon).

44

Atherothrombosis underlies the majority of cardiovascular events independently of

the specific vascular bed in which they occur.(48) Indeed, a high percentage of

atherothrombotic diseases occur in more than one area of the vasculature and, therefore, are

classified as coronary, cerebrovascular or peripheral arterial disease. Data from the

REACH(46) and CAPRIE(47) trials, represented in figure 3, explains the prevalence and

polivascular disease presentation of atherothrombotic disease. The proportion of PAD patients

with CAD and/or cerebrovascular disease (CVD) is 61%. CAD and CVD constitute the two

leading causes of death worldwide,(49) followed by PAD, that is present in a 38.1% of the

total amount of CVD patients.

Cardiovascular disease (CD) is the clinical manifestation of atherothrombosis,

representing 17.9 million deaths globally (39%) and, consequently, a global health problem

at present.(26) The number of deaths caused by CD is expected to reach 23.3 million by 2030

and, thus, CD is projected to remain the leading cause of death worldwide.(50)

Vascular calcification

Calcium is an essential ion in many metabolic pathway and is a widespread cellular

signal effector. It is involved in the thrombosis cascade, the regulation of muscular

contractility (including myocardium), neuronal activity, the endocrine system, and,

paradoxically, in the genesis of vascular calcification (VC). Thus, calcium may accumulate in

the spleen, liver, kidney, and the circulatory system, where it is deposited in the arterial intima

and media layers and may eventually lead to obstruction.(51)

VC is a pathologic response to injuries or toxic stimuli involving inflammatory

cells.(52) Historically, VC was considered to be a passive process, resulting of calcium (Ca2+)

Figure 3. Cardiovascular disease prevalence and presentation. Prevalences of cardiovascular

diseases observed in the REACH trial(46) and percentage of atherothrombotic events depending on

the vascular territories affected. Data obtained from the CAPRIE trial.(47)

45

and phosphate (P) ions exceeding solubility in tissue fluid, thereby inducing the precipitation

and deposition of hydroxyapatite crystals. Far from being a passive process related to aging,

it is an active and, maybe, potentially reversible process.(53,54) VC formation is now

considered a complex, actively controlled intracellular molecular process. It involves the

differentiation of macrophages and vascular smooth muscle cells (VSMCs) into osteoclast-

like cells, similar to that which occurs in bone formation.(55–57) In fact, actual bone structure

and bone related molecules and cell types have been found in calcified lesions.(58) The

oxidative stress caused by locally generated hydrogen peroxide (H2O2) promotes the

metaplasia of VSMCs in the vascular wall to an osteogenic phenotype. Further, this is

associated with a significant loss of endogenous VSMC calcification inhibitors (matrix Gla

protein, a calcium-binding protein involved in bone formation, pyrophosphate, and the

inducible inhibitor osteopontin) and circulating inhibitors, such as fetuin-A.(55) To sum up,

pathophysiological mechanisms resulting in VC can be broadly described as: (1) elevation in

serum Ca2+ and P levels, (2) the induction of osteogenesis, (3) the inadequate inhibition of the

mineralization process, and (4) the migration and differentiation of macrophages and VSMCs

into osteoclast-like cells.(52,55,59)

In diabetic patients, advanced glycation end-products might promote mineralization

of pericytes, and tight glycemic control might slow calcification in type 1 (but not type 2)

diabetes. Moreover, the transcription factor proliferator-activated receptor gamma might deter

calcification by inhibiting Wnt5a-dependent signaling in VSMCs.(60)

As previously mentioned, through the mechanisms of calcification there is an

accumulation of Ca2+ and P in arteries with mineral deposits in the intima or medial layer,(51)

leading to two different types of VC: intimal and medial (or Mönckeberg’s disease).(55)

Intimal calcification is associated with atherosclerotic plaques and thought to result

from modified lipid and low-density lipoprotein (LDL) particle accumulation, pro-

inflammatory cytokines, and apoptosis within the plaque that induce osteogenic cell

differentiation. Calcification in atherosclerosis alters plaque stability by direct mechanical

stress within the fibrous cap and by increasing arterial stiffness and promoting

hypertension.(53,60,61) Perhaps, intimal calcification is the organism’s natural response to

isolate the atherosclerotic plaque and interrupt the progression of abnormal cellular processes,

thereby protecting the healthy adjacent intima.(55) As such, some studies have shown that

calcification, specifically in the coronary bed, plays a major role in plaque stabilization.(62)

In other territories, such as the carotid artery (where the main complications are caused by

embolization rather than acute local occlusion), some studies associated the calcification

46

processes with plaque stability,(63) and other studies correlated a higher load of calcification

with more hemorrhagic complications, suggesting that intraplaque calcification may not

contribute directly to plaque stabilization.(64) Last but not least, it is worth mentioning that

intimal calcification adopts a different composition depending on the territory affected.

Specifically, femoral arteries are usually the most calcified vessels, presenting with advanced

calcifications (sheet-like, nodule) and frequent osteoid metaplasia. In contrast, the plaques in

carotid arteries display an increased lipid content.(53,65–67)

On the other hand, the medial arterial calcification (MAC) is considered to be more

widespread in the lower abdominal region.(55) It is also known as Mönckeberg disease; since

it was firstly described by Mönckeberg in 1903.(55) This process results from the osteogenic

differentiation of VSMCs within the medial layer of the vessel wall.(68) Ca2+ accumulation

begins as an amorphous mineral deposit and undergoes progressive remodeling, ultimately

mineralizing into mature bone. Although medial calcification is generally not associated with

luminal obstruction, the decrease in vessel wall elasticity and increase of its thickness

ultimately lead to lumen loss and reduced perfusion, specially in the microvascular bed.(51)

Some conditions that promote VC are diabetes mellitus (type 1 and 2), ESRD,

hyperphosphatemia and secondary hyperparathyroidism (SHPT). The latter is a condition

associated with progressive renal failure, characterized by increased phosphate levels and

diminished Ca2+, which causes more Ca2+ to be taken from the bones and reabsorbed by the

intestines and kidney. Subsequently, management of SHPT is central to the achieve a

reduction of hyperphosphatemia and hypocalcemia without producing hypercalcemia. This

could be achieved with oral phosphate binders, active vitamin D analogs and Ca2+

mimetics.(51) Phosphate binders (e.g., sevelamer; RenvelaTM) significantly decrease serum

Ca2+ and are considered responsible for the lower rates of VC.(69) Concomitantly, a

significant reduction in aortic calcification was observed in mice with CKD and SHPT under

treatment with vitamin D receptor agonists.(70) In patients with CKD, vitamin D therapy

decreases serum PTH levels, significantly reduces the incidence of cardiovascular events and

improves survival.(71) In patients with ESRD and SHPT, treatment with cinacalcet

(SensiparTM), which controls Ca2+ homeostasis by regulating the release of PTH, results in

fewer hospitalizations for cardiovascular complications compared to placebo.(72) Cinacalcet,

in combination with low-dose vitamin D, also attenuates coronary and aortic calcification in

hemodialysis patients.(73) Figure 4 sums up the different mechanisms and factors implicated

in VC.

47

Figure 4. Diagram of mechanisms leading to vascular calcification.(51)

Arterial calcification is a strong risk factor and independent predictor of

cardiovascular complications and mortality,(74,75) specifically affecting arterial stiffness in

the case of MAC, as described by London et al.(76) Rennenberg et al. studied VCs among

218,080 subjects and found an increased risk (odds ratio, 95% CI) of 4.62 (2.24-9.53) for all-

cause mortality, 3.94 (2.39-6.5) for cardiovascular mortality, 3.74 (2.56-5.45 for coronary

events, 2.21 (1.81-2.69) for stroke and 3.41 (2.71-4.3) for any cardiovascular event.(77) VC

is also associated with PAD, CKD and DM.(51)

From a radiological point of view, arterial intima atherosclerosis produces a patchy

discontinuous appearance of large calcific deposits, while calcification in MAC is more

diffuse and usually affecting the whole circumference of the vessel, adopting a “railroad track”

pattern.(75) These findings can be seen on plain x-ray (figure 5) imaging and are well

Figure 5. Diagram of X-ray appearance of MAC (a) and arterial

intima calcification (b).(79)

48

correlated with microscopic findings.(78)

MAC is found in 11% of routine pelvic x-rays(80) and is commonly found in diabetic

elder males with high cholesterol levels.(81) It is also associated with trophic foot ulcers and

PAD,(82) more severely worsened in the peroneal and tibial arteries than in the thigh arteries.

However, MAC seems to be related to diabetes and its microvascular manifestations rather

than PAD itself.

Smith et al.(79) reported a good predictive value of MAC for podiatric care

requirements and for diabetes. MAC has a positive predictive value of 92.9% (95% CI: 69.2-

98.7) for diabetes and a specificity of 99.9% (95%CI 99.4-100).

It has been reported that MAC has an OR (95% CI) for mortality of 1.5 (1-2.1), for

amputations 5.5 (2.1-14.1), for proteinuria 2.4 (1.3-4.5), for retinopathy 1.7 (0.98-2.8) and for

CAD 1.6 (0.48-5.4). Other studies found that MAC had an OR of 4.2 for cardiovascular

death.(83) There’s also correlation between MAC and other microvascular complications of

diabetes, such as decreased vibration perception, serum creatinine and autonomic

neuropathy.(84) The small vessels are the common factor among all previous clinical effects

of MAC, suggesting a relationship between MAC and microvascular disease.

1.2. Present classifications and risk stratification systems of PAD

1.2.1. Leriche-Fontaine classification

Described in 1954, the Leriche-Fontaine classification is divided into four stages. The

first one (stage I) is defined by demonstrable PAD by signs or explorations but with no

symptoms. Stage II corresponds to intermittent claudication (explained below), subdivided in

IIa as claudication of more than 150 m (non-disabling); and IIb, which affects the quality of

life and supposes a limitation for the patient. CLI is confined to the last two stages: III and IV,

both with the eventual presence of rest pain. In stage IV is also defined by the presence of an

ulceration or gangrene in the foot or the limb.(85)

1.2.2. Rutherford classification

Firstly described in 1986(86) and afterward reviewed in 1997(87), the classification

described by Rutherford et al. had the aim of ease description and to compare patients when

reporting cases and studies of peripheral arterial occlusive disease in the literature. They

described stages in acute and chronic limb ischemia but also the criteria to describe outcomes,

49

such as limb salvage, change in clinical status, patency or risk factors. The classification used

to classify the stages of chronic limb ischemia is shown in Table 1.

Table 1. Rutherford classification. AP: Ankle Pressure; PVR: Pulse Volume Recording; TE: Treadmill

Exercise, consisting of 5 min at 2 mph on a 12% incline; TP: Toe Pressure. *Grades II and III correspond

to critical limb ischemia. (87)

Symptomatic disease is stratified into six categories. Changes between categories

correspond to clinical improvement after treatment. Gangrene is divided into two categories:

affecting only the toes or with a larger tissue loss that could avoid the salvage of a functional

foot remnant.

The zero grade is used to identify patients with no or little symptoms or sensations,

valuable to allow a postoperative gauge at all levels.

Claudication is defined as an extremity pain or weakness produced by a constant

amount of walking and that it is promptly relieved by stopping the muscular activity.

Recommending patients to undergo intervention when disabled is enough for the clinical

practice, but not enough precise for trials since disability is relative to age, occupation among

other factors. The non-invasive vascular laboratory test criteria to define claudication higher

than mild is to not being able to complete 5 minutes on the treadmill at 2 mph on a 12% incline;

with a reduction in ankle pressure to ≤ 50 mmHg, thus corresponding to moderate or severe

claudication. If only some of these criteria are accomplished, the stage corresponds to

moderate claudication.

The grade II or category 4 in this classification corresponds to the ischemic rest pain

and indicates diffuse foot ischemia. It cannot be controlled by analgesics and it is usually

localized to the forefoot or near focal ischemic lesions, increasing at night or when lying down.

The ankle pressure is commonly lower than 40 mmHg, and toe pressures ≤ 30 mmHg.

50

At the last grade, corresponds to cases with gangrene and depending on whether it is

focal - usually to the toes, corresponding to category 5 - or more extended proximally for

category 6. The upper limit of ankle pressure for this grade is 60 mmHg, a toe pressure of 40

mmHg or barely pulsatile tracing at the plethysmography recording.

The concept of critical limb ischemia corresponds to categories 4, 5 and 6 in this

classification, thus defined by rest pain, nonhealing ulcer or gangrene besides the evidence of

distal ischemia.

1.2.3. University of Texas ulcer classification

The goal of Lawrence et al.(88) from the Department of Orthopaedics, University of

Texas Health Science Center, San Antonio defined this classification ultimately known as

TUC classification. This system to improve communication, leading to a less complex, more

predictable treatment course and, ultimately, improved results. The following classification

uses a system of wound grade and stage to categorize wounds by severity. Wounds are graded

by depth. Grade 0 represents a pre- or post-ulcerative site. Grade I ulcers are superficial

wounds through the epidermis or epidermis and dermis but do not penetrate to tendon, capsule,

or bone. Grade II wounds penetrate to tendon or capsule. Grade III wounds penetrate to bone

or into a joint. Within each wound grade, there are four stages: non-ischemic clean wounds

(A), non-ischemic infected wounds (B), ischemic wounds (C), and infected ischemic wounds

(D). Table 2 is the original table for the classification.

Table 2. University of Texas Ulcer Classification (TUC), reported as the grade followed by the

stage names.

51

1.2.4. TASC anatomical classifications

The Transatlantic Inter-Society Consensus (TASC II) for the Management of Perip-

heral Arterial Disease, has provided expert recommendations on the diagnosis and treatment

of PAD. One of the main used aspects of these guidelines is the TASC artery lesion classifi-

cation. It is based on anatomic criteria and provides a characterization of the patterns of di-

sease and suggest treatment decisions. Based on the complexity and anatomical location of

the lesions, recommendations tend to endovascular or open surgical repair.

TASC anatomical descriptions of the lesion patterns enable comparison between various

grades of complexity throughout three territories: the aortoiliac, the FP and the BTK seg-

ment. The lesions are classified as A, B, C or D. Despite simpler lesions (A and B grades)

are considered more suitable for endovascular repair and the more complex ones (C and D)

for open surgery; TASC B and C lesions treatments could be considered with other factors

such as the technical resources, patient status, and the surgeon’s experience. they are not

sufficient, anyway, to guide clinical decisions because of a lack of randomized clinical trials

to generate particular recommendations.

Although the aortoiliac lesions are not concerned in this project, the aortoiliac classification

worth a mention as a part of the TASC classification. As it is shown in Figure 6, the grades

are: (15,89)

- TASC A: Common iliac artery stenosis or external iliac stenosis of ≤ 3 cm

- TASC B: Stenosis of infrarenal aorta of ≤ 3 cm, common iliac artery occlusion, ex-

ternal iliac stenosis of ≤ 10 cm or unilateral occlusion.

- TASC C: Bilateral common iliac artery occlusions, bilateral external iliac stenosis

of ≤ 10 cm or unilateral stenosis extending to the common femoral artery, unilateral external

iliac occlusion involving the internal iliac artery or unilateral heavily calcified occlusion of

external iliac.

- TASC D: Infrarrenal aortoiliac occlusion, stenosis of the whole aortoiliac segment

or diffuse pathology affecting the common femoral artery, bilateral external iliac occlusions

or iliac stenosis in patients with aortic aneurysms.

52

Figure 6. Aortoiliac TASC classification In the FP sector, the criteria for each grade of the classification are (Figure 7):(15,89)

- TASC A: single stenosis ≤ 10 cm or occlusion ≤ 5 cm in length.

- TASC B: multiple lesions (stenosis or occlusions) ≤ 5 cm each, single lesion ≤ 15

cm (not involving infra-geniculate popliteal artery) or heavily calcified occlusion ≤ 5 cm.

- TASC C: multiple lesions totaling ≥ 15 cm or recurrent stenosis after failing treat-

ment.

- TASC D: chronic total occlusions of SFA ≥ 20 cm or popliteal artery and proximal

trifurcation vessels.

Figure 7. Femoropopliteal TASC classification

Finally, the 2015 update on TASC classification (Figure 8), mentioning the below-

the-knee sector is defined by the lesions on the target vessel (assuming worse stenosis or

occlusions in the other arteries):(89)

- TASC A: stenosis ≤ 5 cm in length

- TASC B: multiple stenosis of ≤ 5 cm each with a total length ≤ 10 cm or a single

occlusion ≤ 3 cm in length

- TASC C: multiple stenosis or a single occlusion of > 10 cm in length

53

- TASC D: the same lesions of TASC C but with dense lesion calcification or no

visualization of collaterals

Figure 8. Below-the-knee TASC classification

Specific considerations on the application of the classification to the patient’s

population of the study are explained further in the discussion section.

1.2.5. SVS Wound Ischemia foot Infection (WIfI) stages(1)

First defined in 1982, CLI was intended to refer to patients with a threatened lower

extremity because of chronic ischemia, but excluding diabetic patients whose should be

analyzed separately.(4) Nevertheless, among patients with current accepted criteria for CLI,

diabetes is highly prevalent. This liberal application of the term CLI could explain the issues

in measuring and comparing outcomes of treatment options, especially as treatment options

have rapidly expanded.

To provide a more precise description of the disease burden to allow accurate

outcomes assessments and comparisons of various treatments in CLI patients, the Society for

Vascular Surgery (SVS) Lower Extremity Guidelines Committee undertook the task of

creating a classification taking into consideration the whole clinical scenario of CLI in

diabetic patients. Also, it is suggested that an updated comorbidity index and a simpler

anatomic classification system will need to be added to the classification. Not only to stratify

the disease burden but to aid in the selection of the best therapy for any given patient.(1)

Thus, in addition to the perfusion, the extent of the ulcer and the degree of infection

are weighted in this classification: Wound, Ischemia, and foot Infection (WIfI). Each of the

three factors (or component) is graded on an increasing scale from 0 (none) to 3 (severe). The

thresholds for the grades are the following:

54

- Wound:

o 0: no ulcer

o 1: shallow ulcer, without exposure of deep tissues

o 2: bone, joint or tendon exposures

o 3: extensive and deep ulcer

- Ischemia:

o 0: ABI ≥ 0.8, Ankle pressure > 100 mmHg, TP or TcPO2 > 60 mmHg

o 1: ABI 0.6 – 0.8, Ankle pressure 70 – 100 mmHg, TP or TcPO2 40 – 60

mmHg

o 2: ABI 0.4 – 0.6, Ankle pressure 50 – 70 mmHg, TP or TcPO2 30 – 40

mmHg

o 3: ABI < 0.4, Ankle pressure < 50 mmHg, TP or TcPO2 < 30 mmHg

- Foot Infection:

o 0: No symptoms or signs of infection

o 1: Local infection (after excluding an inflammatory response of the skin)

involving only the skin and the subcutaneous tissue.

o 2: Local deep infection or with erythema >2 cm, with no systemic

inflammatory response.

o 3: Local infection (as in grade 2) with the signs of SIRS, as manifested by

two or more of the following: Temperature >38º or <36ºC, heart rate >90

beats/min, respiratory rate >20 breaths/min or PaCO2 <32 mm Hg, white

blood cell count >12,000 or <4000 cu/mm or 10% immature forms.

The set of three grades on each component is used to define the class. The classes

arranged in tables (Figure 9) and using a Delphi Consensus from the opinion of a group of

experts, one of the 4 possible stages were assigned: very low (VL), low (L), moderate (M) or

high (H). The stages were defined for both the estimate amputation risk and the likelihood of

benefit of revascularization.(1) This system had been after validated several times by other

independent studies.(18–20,22–25)

55

Figure 9. WIfI consensus classification describing (a) the estimated risk of amputation at 1 year and

(b) the estimated likelihood of benefit of revascularization.

1.3. Present methods for ischemic assessment

The initial assessment of PAD and CLI patients a direct inquiry on previous

myocardial infarction, stroke, coronary artery bypass graft (CABG) and arterial surgery

should be included. It is not exceptional that polivascular disease to be diagnosed for the first-

time during PAD examination. An inquiry for cardiovascular risk factors is equally important.

Upper extremity exertional pain, postprandial abdominal pain, and erectile dysfunction are

other important potential clues in the medical interview.(90)

Physical examination should be systematic and include blood pressure in both arms,

palpation of all peripheral and central pulses, notifying any thrill or murmur on them; pulse

rate, rhythm and heart sounds; examination of the hands and abdominal aorta. On the feet, it

is mandatory a careful examination, between the toes, assessing skin integrity, color,

temperature and calf hair loss. The palpation of pulses is subjective and is influenced by the

sensitivity of the fingers, the experience of the examiner, the obesity and edema of the patient

and the warmth of the room. The femoral artery, whether pulsatile or occluded, should always

be palpable unless the patient is very obese.(90)

56

In patients presenting with classic history of IC but with palpable foot pulses, there is

usually proximal aortoiliac disease with collaterals through the pelvis, which produce

adequate blood flow at rest but decreases suddenly with a small exercise like walking up and

down the corridor or on a treadmill, or even by a repeated “tiptoe” while leaning on the

couch.(90)

1.3.1. Ankle Brachial Index and ankle pressure.

Measuring the pressure in the ankle arteries has become a standard part of the initial

evaluation and differential diagnosis of patients with suspected PAD. The ankle-brachial

index (ABI) value raises from the ratio between the systolic pressure measured in the calf (for

the peroneal, posterior and anterior tibial arteries) and the higher brachial pressure of either

arm. The index leg is often defined as the leg with the lower ABI.(15) The ABI is cheap and

non-invasive, which makes it potentially valuable in health care. A reduced ABI is a potent

predictor of the risk of future cardiovascular events.(91)

The typical threshold for diagnosing PAD is ≤ 0.90 at rest, with an upper bound in

normality from 1.2 to 1.4. An ABI <0.5 is often associated with critical limb ischemia. If we

consider only the ankle pressure in patients with ischemic ulcers, it is typically 50–70 mmHg,

and in patients with ischemic rest pain it stays around 30–50 mmHg.(15)

The ABI can represent a wide range of ankle, depending on the systolic actual pressure.

For example, and ABI of 0.3 with a systolic blood pressure of 110 or 160 mmHg are 33 and

48 mmHg respectively, and near both opposite bounds in the ischemic pain threshold.

However, ABI is better for comparing groups of patients or for monitoring a given patient

before and after eventual treatments.(87)

However, the ABI is not a useful test for detecting PAD in patients with diabetes and/or

renal insufficiency.(92) Due to vascular calcification, the tibial vessels at the ankle become

non-compressible. This leads to a false elevation of the ankle pressure, in which ABI it

typically >1.40. In these patients, additional non-invasive diagnostic testing should be

performed.(15)

1.3.2. Toe pressure

Toe pressures (TP) have main importance in diabetic patients because it is the most

distal pressure representing the big vessel circulation that can be measured, and the toe arteries

are less prone to MAC.(93) The TASC II threshold for ischemia symptom appearance is below

50 mmHg.(15) In addition, recent European guidelines on diabetic foot management have

57

raised the level of probable wound healing from the previous 30–50 mmHg to 50–55 mmHg.

In non-diabetics, the cut-off value for rest pain has been set on 30 mmHg.(93,94)

Few studies have shown the usefulness of TP compared with ABI in predicting

outcomes of CLI patients.(95,96) This may be secondary to the predominance of small-vessel

disease in the foot among patients with CLI. TPs are also less likely to give falsely elevated

pressures than are APs.(16)

TP could be measured by three methods: the mercury strain-gauge,

photoplethysmography (PPG) and laser Doppler (LD), the latter two being the ones still in

use. As it is mentioned, PPG is based on detecting changes volume of the toe by emitting

infrared light that penetrates the tissue under the probe. The LD emits infrared laser light that

is reflected from moving particles (red blood cells). The PPG detectors need a pulsatile flow,

whereas LD sensors detect minor flows; thus, LD allows lower values of toe pressure to be

measured.(93) This test, as the TcPO2, is sensible to the vasodilatation due to the temperature

of the ambient and probe, and it should be standardized. It is a quick test and thus suitable for

everyday practice.(93)

1.3.3. Pulse volume recordings

Pulse volume recordings (PVR) provide a qualitative measurement by inflating cuffs

at specified levels on each lower extremity. The cuff measures the minuscule change in the

volume of the limb with each pulse, creating a volume/time trace. PVRs are useful in patients’

extremely calcified vessels, as this modality relies on limb volume change rather than the

compression of the vessels. Unfortunately, as it is mentioned, it is only a rough estimate of

the level of disease, without correlation with the severity or prognosis of the disease.

Nonetheless, PVRs can provide information about changes in arterial flow quality between

segments of the main vessels of the limb, giving better anatomical information than other

techniques.(97)

1.3.4. Photoplethysmography

Photoplethysmography (PPG) uses a transmitted infrared signal through each of the

digits. The degree of transmitted signal varies depending on blood volume within the digit,

blood vessel wall movement, and the orientation of red blood cells.(98) PPG is useful for the

detection of disease below the knee as well as disease isolated to the forefoot and digits. The

qualitative restoration of pulsed wave within previously “flat-lined” tracings can be assumed

as an improvement of distal perfusion or suggesting restenosis when losing the pulsed

58

wave.(97) Anyway, PPG is only comparable with a previous reference on the same patient

and used as subjective information. The role of PPGs in the surveillance of patients with distal

ulcers is essential and may provide information about any restenosis of a distal pedal

intervention.

1.3.5. Transcutaneous oxygen pressure (TcPO2)

Toe pressure and TcPO2 measurements are the most widely used noninvasive methods

in assessing foot perfusion and wound healing potential.(93) The TcPO2 measurement allows

the determination of the amount of oxygen that has diffused from the capillaries, through the

epidermis, to an electrode sensor at the measuring site, usually placed at the dorsum of the

foot.(99) The patient should be in supine position in a room maintained at 22ºC and the

electrode at 45ºC. Finally, calibration time from 10 to 30 minutes is needed prior to the

continuous measure. (99,100)

An improvement in the TcPO2 value postintervention compared with preintervention

has been validated as an excellent marker of tissue reperfusion. TcPO2 values greater than 40

mm Hg in the area surrounding the ulcer or amputation site are considered predictive of

successful healing. This test has the advantage of not being limited by non-compressible

vessels, and in patients without pedal Doppler signals, it is one of the noninvasive tests that

are helpful to assess perfusion.(100) Another study in diabetic patients made by Kalani et

al.(101) found that the ulcers on the feet of patients with TcPO2 > 25 mmHg healed within

4–6 weeks; while when the TcPO2 was < 25 mmHg the ulcers did not heal (sensitivity: 85%;

specificity: 92%). The TcPO2 values proposed by the TASC II(15) for a critical level is 30

mmHg. In figure 10(102) the probability of DFU healing is represented, according to the

TcPO2, the TP and the AP. In all the three measures there is a range of medium probabilities

for healing, being conflictive the choosing of a certain threshold.

Despite TASC II recommendations for the use of TcPO2 in all patients with CLI, the

use of this diagnostic tool remains low related to its limited availability in the office setting,

patient resistance to avoiding smoking and caffeine before the test, as well as the time and

cost constraints associated with providing the temperature-controlled environment required

to standardize the test.(99)

59

Figure 10. Probability of ulcer healing as related to different levels of systolic ankle pressure, toe

pressure and TcPO2 (from the International Consensus of the Diabetic Foot 1999).

1.3.6. Tissue oxygen saturation mapping

On a similar basis than in the conventional TcPO2, a team of Kagaya et al.(103) used

a near-infrared tissue oximeter monitor to detect the ischemic areas in the entire foot of a CLI

patient. They measured the saturation of oxygen (StO2) of many points on the foot to make a

map. The feasibility of the test relies on that the StO2 values can be measured within only10

seconds per point, by contrast to the conventional TcPO2. The contribution of this study is

that the angiosomes are more variable and patient-dependent, likely due to collateralization

and variations in microcirculation. This interesting fact could provide useful information to

determine whether further interventions should be done and monitoring personalized

treatment goals for each patient.

1.3.7. Hyperspectral imaging

Hyperspectral imaging (HI) consists of recording a series of images representing the

intensity of diffusely reflected light from biological tissue at discrete wavelengths. The

resulting set of images is called hypercube denoting spatial coordinates (x, y) and a spectral

coordinate (λ). The analysis of the hypercube asses the distribution of chromophores, such as

melanin or hemoglobin. That means that we can estimate the hemoglobin concentration and

oxygen saturation from the tissue’s diffuse reflectance.(104)

60

HI has been used to determine the spatial distribution of oxygen saturation in human

skin(104,105) and to detect the changes in the diabetic foot.(106–108) Nouvong et al.(106)

evaluated prospectively the predicting potential of HI for healing of ulcers in 66 diabetic

patients (type 1 and 2). The sensitivity was 80% (43 of 54), the specificity was 74% (14 of

19), and the positive predictive value was 90% (43 of 48). A reason for false-positive was

attributed to osteomyelitis and, on the other hand, callus could explain some of the false-

negatives. For this study, the averaged oxyhemoglobin (TOXY) and deoxyhemoglobin (TDEOXY)

concentrations, as well as oxygen saturation over an approximately 1-cm-thick band within

the peri-wound area, were calculated around ulcers and compared between those healing and

those non-healing. Figure 11 shows an example of the HI color coding in a peri-wound

tissue.(106)

Figure 11. Hyperspectral imaging image sample. Visible (left) and hyperspectral (right) images of a

healing DFU. The top panels show a healed DFU with oxy, deoxy and StO2 values of 75, 34 and 69%,

respectively. On the other hand, the bottom panels show a non-healed DFU, where the abovementioned

values are 60, 53 and 53%.(106)

Using the same HI dataset, Yudovsky et al.(104) explored the risk of developing an

ulcer in diabetic foot using HI images of diabetic feet collected before ulceration and dividing

groups from the combined values for oxyhemoglobin (TOXY = 18) and deoxyhemoglobin

(TDEOXY = 5.8). They found that diabetic foot ulcer formation could be predicted with a

sensitivity and specificity of 95% and 80%, respectively. Those findings confirmed the results

of a previous pilot study by Khaodhiar et al.(107)

To sum up, HI is noninvasive and can assess oxygen saturation with high spatial

resolution in a relatively short time, with no special preparation.(109) However, HI has some

limitations including the need to know the chromophores and structure of the tissue or even

61

the presence of wounds or scars prior to image analysis.(110,111) However, there are no large

studies that could validate this technique as a reference in clinical guidelines.

1.3.8. Oxygen-to-see (O2C) method

The oxygen-to-see (O2C) method (LEA Medizintechnik GmbH, Gießen, Germany) is

an optical measuring technique that uses both white light spectrometry and laser Doppler

flowmetry. This combination allows measuring three parameters at a time: oxygen saturation

(SatO2), relative hemoglobin (rHb) and blood flow.

Figure 12. O2C measuring principle is a combination of white light spectrometry and laser

Doppler flowmetry (courtesy of LEA Medizintechnik GmbH).

The method is based on the reflection from the skin of a broadband light signal (500–

850nm) as well as light from a laser source (830nm). The white light source spectrum is used

to determine SatO2 and rHb and the laser light determines blood flow (Figure 12). The

penetration depth depends on the selected probe and it could be up to 8mm.(112)

The test showed high sensitivity and specificity for predicting the healing potential in

above the ankle amputations.(113) This technique can also be used for periprocedural

measurements in the foot. The probe can measure perfusion in a concrete angiosome (as

shown in figure 13) after or during the revascularization. Indeed, several studies have

demonstrated an increase of foot perfusion independently of the direct angiosome

revascularization.(114,115) However, no large studies with clinical follow-up are available to

validate this technique in clinical practice.

62

Figure 13. The O2C probe, as

placed on several angiosomes.

1.3.9. Micro Oxygen sensors (MOXYs)

Performed by Montero-Baker et al.(116), the first-in-man evaluation of a potential tool

that consists in micro-sensors placed in the subcutaneous tissue; from 2 to 4 mm below the

skin surface, and were injected using an 18-gauge injector. They are settled in three locations

of interest (with consideration of the ulcer or wound if present, angiosome anatomy or their

likelihood of presenting a change in oxygenation after the revascularization). The micro-

sensors (0.5*0.5*5 mm) were designed to remain in the body permanently, avoiding

inflammatory reaction by physical and chemical properties. They were soft and tissue-like to

minimize stress at the material-tissue interface caused by motion and pressure, which can

damage or stimulate adjacent immune cells and prolong the inflammatory phase. From that

study, 97.2% of sensors were successfully located and measured and the increase in median

oxygen concentration measured in the treated feet was statistically significant (p=.0042),

while arm sensors showed negligible or decreasing oxygen changes.

1.3.10. Indocyanine green fluorescence imaging

Fluorescence imaging consists in the injection, either intravenous (at the bedside or

in the clinic) or intra-arterial (during an EVT, for instance), of indocyanine green (ICG) to

quantify tissue perfusion (figure 14). The protocol for the intravenous technique consists in

placing a camera at 20 cm over the foot while the patient is in supine position after at least

15 minutes of rest, with a room temperature between 20 and 25ºC removing dressings and

extinguishing overhead lights. The injection dose of ICG is 0.1 mg/kg of 0.1% ICG intra-

venously administrated solution, and pushed with 10 mL of normal saline to flush the intra-

63

venous route. After that injection, the capturing and recording go-ahead for 5 minu-

tes.(117,118) After the injection, the detector emits low-level light at a wavelength of 760

nm over the skin, stimulating the fluorescence of the IGC molecule. This is captured by the

camera and displayed on the monitor, after assigning a numerical value and therefore tracing

a time/intensity curve. That can be measured in one specific point or within a freehand re-

gion of interest (ROI). The pre- and post-EVT curves can be plotted on the same graph, in

the same way than in the PA.(97)

Figure 14. Fluorescence angiography image measuring surface foot perfusion using an intravenous injection of in-docyanine green (inset: picture of the same foot).

This technique had been used initially to visualize retinal vascularity. It has a very

strong safety record (1 adverse event per 40,000 administrations). IGC rapidly binds to albu-

min, so it remains intravascular for prolonged periods, making it an ideal marker to measure

the perfusion. The hepatobiliary excretion makes it safe for CRD patients.(119)

A study made by Igari et al.(117) compared pre- and post-EVT ICG fluorescence imaging

(ICG-FI) on Rutherford 2-4 patients. They found that a time from fluorescence onset to half

the maximum intensity (T½) >20 seconds for predefined ROI (which included the distal re-

gion of the first metatarsal bone) was significantly correlated with a toe pressure of <50

mmHg (sensitivity: 0.77, specificity: 0.80).

Another study made by Venermo et al.(118) was focused on CLI, with 66% of dia-

betic patients and 70% had an ischemic tissue lesion. They measured the ICG-FI in each leg.

Figure 15. Time-intensity

curves from ICG-FI. (cour-

tesy of Rother et al.)

64

They measured two parameters, the T½ and the PDE10 (defined as the increase of the inten-

sity during the first 10 s). Time-intensity curves, with the aspect that is shown in figure 15,

were repeatable and there was a strong correlation between TcPO2 and PDE10 was strong

in diabetic patients (R=.70, p:.003). Other studies have confirmed the reliability and clinical

correspondence with the information from the ICG-FI.(120) Anyway, the application of this

technique become problematic in patients with ulcers and concomitant foot infection. In

such cases, imaging could tend to overestimate perfusion.(121)

1.3.11. Peripheral duplex ultrasound imaging

Duplex ultrasound (DUS) is based on a gray-scale 2D imaging, color doppler and

spectral waveform analysis. It has been a mainstay of vascular imaging for decades and it is

still recommended as the first-line imaging study to the follow-up and even the intervention

planning. It is non-invasive, yields true hemodynamic information, it does not affect the renal

function and it is relatively inexpensive.(93)

There are several criteria to determine any stenosis as significant, besides of the

anatomical and morphological information of the B-mode.(93) First, color doppler change

from a smooth and uniform color associated with laminar flow to that of mixed color image

associated with turbulence at and distal to the examined lesion. On the other hand, the form

of the spectral doppler, from the normal triphasic or biphasic pattern with narrow spectral

width to that of a monophasic tracing with spectral broadening distal to the lesion. Finally, the

Peak Systolic Velocity (PSV) and the End Diastolic Velocity (EDV) are the more precise

measures: a PSV of >200 cm/s and a ratio between two segments >2.0 translate a

stenosis >50%, and PSV >300 cm/s, EDV 40-100 cm/s and a ratio >4.0 means a severe

stenosis (>75%).(122)

The accuracy of DUS has been compared with the gold-standard, the Digital

Subtraction angiography (DSA). Several meta-analyses(123) have shown high sensitivity

(84%–91%) and specificity (93%–96%) DUS compared to DSA.

Moreover, there are some other advantages of this technique; like inquiring about the

availability of saphenous vein for bypass grafts, the presence of a popliteal artery aneurysm

or choosing the most useful outflow artery for distal bypass. DUS has a special interest due

to its high sensitivity for the detection of patent tibial arteries; it is excellent for evaluating in-

stent restenosis or occlusions(124) and the plantar artery or dorsalis pedis arteries when

occluded could not be visible in DSA but can be seen in DUS.(93)

65

The main limitation of DUS exploration and the interpretation of its findings is that it

is very operator dependent since DUS image alone is not informative without a summarizing

cartography.(93) DUS is a more time-consuming examination than CTA and MRA. The

visibility to the distal aorta, pelvic area, and the peroneal artery may be limited, added to the

shadows caused by calcification, the reliability of the exploration could be diminished.(124)

Moreover, DUS is less accurate when exploring arteries distal to occlusions or tight stenosis,

where the velocity curve is not diagnostic due to slow velocity.(93)

1.3.12. Tomographic ultrasound imaging systems

Multispectral optoacoustic tomography (MSOT, iThera Medical GmbH, Munich,

Germany) and other tomographic imaging systems based on ultrasounds suppose novel and

promising techniques of noninvasive perfusion measurement. This technique combines

conventional ultrasound with pulsed laser light in the near-infrared range. The laser pulses

cause an extremely slight and transient rise in temperature in the tissue. This results in

thermoelastic expansion, which generates an ultrasonic wave detected by highly sensitive

detectors. Light excitation at specific wavelengths enables conclusions to be drawn on the

composition of the tissue based on the absorption spectrum of several molecules (e. g.,

oxygenated Hb, deoxygenated Hb, melanin, lipids or contrast agents such as indocyanine

green or methylene).(112,125)

Similarly, Plumb et al.(126) tested another method: photoacoustic imaging using a

Fabry-Perot interferometer-based device which generates three-dimensional images of

human vasculature at a depth of 14 mm. It was sensible to vasomotor microcirculatory

changes induced by thermal stimuli, suggesting that it can provide valuable, objective and

precise information about blood perfusion in the foot.

1.3.13. Computerized tomography angiography

Computerized tomography angiography (CTA) is comparable to the gold standard

(catheter-based angiography) in detecting hemodynamically significant stenosis (>50%) with

sensitivity, specificity, and accuracy of 99%, 98%, and 98%, respectively.(127) The main

limitation relies on the below-the-knee (BTK) distribution, where the contrast in the small

lumen of the tibial vessels can be difficult to make a distinction to the mural calcification.

This often leads to lacking accuracy in diagnosis.

66

Dual-energy CTA is a technology able to subtract calcified plaque and soft tissues

from contrast, creating CTA datasets and angiogram-like images. No detectable mean

variation when analyzing vessel segments were found between dual-energy CTA compared

with DSA, with less radiation than it used to be.(128)

1.3.14. Magnetic resonance imaging angiography

Magnetic resonance imaging (MRI) and CTA accuracy have become nearly equivalent

in the below the groin segment., with a slight edge for MRI in the BTK distribution. Newer

protocols, such as the time-resolved imaging of contrast kinetics (TRICKS) allows selecting

the time point at which each segment of the vessel tree is optimally opacified. This technique

increases sensitivity, specificity, and accuracy, particularly in CLI patients.(129) Time-of-

flight (TOF) is also capable to avoid contrast use, but unfortunately, it is prone to artifact with

nonlaminar flow typical of atherosclerotic plaque.(130)

More recently, magnetic resonance perfusion imaging using arterial spin labeling has

been used to quantify arterial flow in the thigh and calf musculature, which has been shown

to have equal or greater sensitivity for PAD compared with ABI.(131,132) Anyway, at present,

both CTA and MRI roles are bounded in planning treatments. (97)

1.4. Natural history of critical limb ischemia

Patient outcomes in CLI are largely determined by morbidity and mortality caused by

cardiovascular events and functional impairment caused by limb loss. Cardiovascular events

such as myocardial infarction and stroke occur in 30-50% of patients with CLI, which face

this risk over a 1-year period, and the same risk is faced over a 5-year period by the whole

spectrum of PAD patients. Similarly, although the risk of major amputation is <5% in 5 years

in patients with IC, it is of 30-50% in the first year in patients with CLI who are not

revascularized.(133)

In the clinical scenario of CLI, revascularization is necessary and an attempt at

treatment is reported in as many as 90% of CLI patients.(15) However, a subgroup of patients

presents a favorable prognosis under conservative management. Marston et al.(134) reported

142 patients with limb ischemia (ABI < 0.7) treated at a comprehensive wound care center

with amputation rates of 19% at 6 months and 23% at 12 months. Another study performed

by Elgzyri et al.(135) evaluated 602 patients with diabetic foot ulcers who had TP < 45 mmHg

or an AP < 80 mmHg. 50% of the cohort healed primarily without revascularization.

67

Nevertheless, besides the relieve of rest pain and/or faster healing rates, limb conservation

with optimal revascularization, if feasible, would outweigh these performance goals.

Prognosis concerning limb salvage and survival in CLI patients has improved over the

years.(136) As such, in large population-based studies performed between 1996 and 2006, a

reduction of major amputation rates from 263 to 188 per 100,000 in Medicare beneficiaries

with PAD (relative risk 0.71; 95% CI: 0.6–0.8) was observed.(137) Similarly, another study

showed that among PAD population over 65 years of age, the adjusted odds ratio of lower

extremity amputation per year between 2000 and 2008 was 0.95 (95% CI: 0.95–0.95,

P<0.001).(138) Consistently, Egorova et al.(139) showed a reduction in surgical interventions

in a US CLI population, where being major amputations decreased from 42% to 30% between

1998 through 2007. Furthermore, while the 1-year amputation-free survival (AFS) reported

in trials performed during the period 1996-1999 was 28-40%, the same outcome from 2006

to 2010 was 48-81%.(140) This trend is supported by a decline in major amputation rates in

CLI patients from 20-50% to 10-38% during the same period. This could be explained by

either an improvement in revascularization techniques and endovascular options available, or

a decrease in CLI incidence due to increased public awareness, better medical therapy and

improved wound care.(136,139,141)

1.4.1 Treatment and management of critical limb ischemia

The clinical presentation of CLI depends on the degree of ischemia, the presence of

infection and neuropathy. The primary goal is to preserve limb function and revascularization

is a fundamental strategy for limb preservation, but in some patients, successful

revascularization is not enough, such as in the case of cognitive impairment or non-

ambulatory status.(130)

A revascularization planning should be made from arterial imaging, both functional

and anatomical. Initial medical treatments include pain control, usually requiring narcotics,

dressing for ulcers and tilting the bed downward to increase limb dependency and

perfusion.(15)

In many centers, endovascular revascularization is the favored approach to CLI

because of lower morbidity and mortality rates than those observed when performing open

surgery. The optimal treatment strategy (endovascular versus open surgery) will depend on

anatomic factors, comorbidities, patient preference, and operator experience and skill.(130)

68

The ESC 2017 guidelines of treatment of PAD(142) propose an algorithm (figure 16) to

evaluate patient candidacy to revascularization.

Open surgery has higher risks of perioperative myocardial infarction, death, and stroke

than endovascular revascularization. However, in CLI without endovascular options or failed

and a reasonable 2-year survival, the potential loss of limb and function may favor

surgery.(143)

Common femoral disease often involves the profunda and superficial femoral artery

(SFA) origins and in this segment. In such cases, surgical endarterectomy with patch

angioplasty offers a more durable result than the endovascular recanalization. The procedure

after the patch angioplasty can be followed with a hybrid endovascular-surgical approach,

which is commonly used in multilevel disease.(143)

Figure 16. CLI treatment algo-rithm, as propo-sed in the 2017 European Society of Cardiology Guidelines on the Diagnosis and Treatment of Pe-ripheral Arterial Diseases, in co-llaboration with the European So-ciety for Vascu-lar Surgery.(142) GSV: great saphenous vein. a) In bedridden, demented and/or frail patients, pri-mary amputation should be consi-dered. b) In the absence of contra-indica-tion for surgery and the presence of adequate tar-get for runoff.

69

Bypass relies on good inflow and outflow, as well as on the conduit type, that could

be autogenous or prosthetic. Autogenous saphenous veins provide better long-term patency

than prosthetic grafts,(144,145) particularly when there is good quality, long, single-segment,

autogenous vein with a diameter of at least 3.5 mm.(146–148) The 3 types of saphenous vein

bypasses are reversed, non-reversed and in situ bypass.

The standard endovascular strategy is a step-by-step approach as shown in figure 17.

An ultrasound-guided femoral antegrade ipsilateral access is often possible and it is more

effective than contralateral retrograde or brachial approaches. The first-line approach, irres-

pective of the length of the lesion, is to cross in an endoluminal position. Nevertheless, in

long calcified occlusions is frequent the need to go on the subintimal space. Despite of all, if

the wire could not arrive at a distal lumen, a bailout strategy is the retrograde access techni-

ques. Those are ideally performed the closer as possible to the lesion that is about to be

treated.(149) In a femoropopliteal occlusion, the retrograde popliteal approaches could be

made in P1 (from the adductor canal to the upper border of the patella) tract or P3 tract

(from the joint line to the emergence of the anterior tibial artery) of the popliteal artery.

When the occlusion is limited to the SFA, the P1 access is enough, but the occlusion or the

subintimal dissection usually arrives at P2 or even P3. In that case could be helpful an ante-

rolateral approach to the P3 tract of the popliteal artery or the tibioperoneal trunk (TPT),

maintaining the patient in supine position.(150) In a CLI scenario, nonetheless, the more fre-

quent affection is in the tibial vessels; thus, the need for a retrograde puncture will be in the

anterior or posterior tibial arteries, dorsalis pedis, peroneal or extreme trans-metatarsal or

trans-plantar arch access.(149,151)

Figure 17. Step-by-step approach to

chronic total occlusions.

70

After crossing the lesion, the usefulness of an atherectomy should be set.(152)

Alternatively, primary balloon angioplasty is made, with or without predilatation. Finally,

flow-limiting dissections or immediate recoiling to > 50% could justify the placement of a

stent. Recently there’s uprising evidence of the lower restenosis rates with the drug-eluting

balloons and stents,(153–156) so we should consider the potential benefit of them on the

patient we are treating. However, in late 2018, Katsanos et al. published a metanalysis

reporting late all-cause mortality of patients in the DCB and DES arms of randomized clinical

trials exploring the treatment of the superficial femoral artery of claudicants.(157) Despite the

safety concern due to the finding, neither a plausible mechanism nor a specific cause of death

were identified.

To take into account the big picture, the only randomized study comparing

endovascular versus open surgical treatment of patients with CLI is the Bypass Versus

Angioplasty in Severe Ischemia of the Leg (BASIL) study.(158) This trial published in 2005,

demonstrated no difference in major amputation or death >5 years. Rates of myocardial

infarction, wound infection, pulmonary complications were higher in the surgical group, and

repeat revascularization was higher in the endovascular arm. However, after the time of the

trial both techniques have improved as well as the perioperative mortality, particularly

concerning the catheter-based techniques.

More recently, several proposals for important endpoints in CLI have converged on

limb salvage (avoiding a major amputation) and the need for major reintervention (a new

bypass graft or thrombolysis of treated segment), and assessments considering the quality-of-

life and cost-effectiveness too.(159,160) As a result of these developments a new randomized

trial of open surgery versus endovascular revascularization in CLI is being performed, the

Best Endovascular Versus Best Surgical Therapy in Patients With Critical Limb Ischemia

(BEST-CLI) study.(161) It has enrolled, by 2017, 775 patients out of 5000 aimed to be

enrolled. It is a timely and critically needed trial that will significantly impact clinical practice

by defining an evidence-based standard of care for patients with CLI.(162)

All efforts in revascularization and maintaining the walking ability usually go through

a minor amputation. Major amputations (below or above the knee), on the other hand, limit

functional self-sufficiency and require prosthesis to walk. Nonetheless, in certain cases, a

primary amputation is the best option in patients with extensive tissue loss or infection, and

in potentially non-ambulating patients.(163)

71

Complementary to all the aforementioned techniques, there are antibiotics and wound

care which aim to improve perfusion, treat the infection, avoid pressure on a

wound, debridement (by scalpel, collagenases, or even maggots(164)) and adequate nutrition.

The perfusion can be increased by increasing local temperature of the limb with sheepskin

boots or by negative pressure dressings (vacuum-assisted).(130)

1.4.2. Endpoints of treatment

The goals of treatment for CLI are to preserve life and limb function by healing the

ulcer and relieving pain while minimizing the frequency and magnitude of interventions.

Despite the scope of the problem, there’s a lack of high-quality evidence to support current

treatment paradigms in CLI. Conte et al.(165) set the Objective Performance Goals (OPG) for

safety and efficacy for catheter-based treatments or endovascular treatments (EVT). However,

the measured variables in the OPG do not say anything about the necessary results that

revascularization should reach, but rather refer to the long-term clinical outcomes of a certain

treatment.

In several reports, 10–15% of CLI patients not considered as suitable for

revascularization heal with no amputation.(102) Thus, besides of risk scores and

classifications, a part of the CLI population will heal whether we do or not the appropriate

revascularization. However, an important tenet in restoring perfusion and blood flow is

establishing in-line flow from the aorta to the affected foot or, more importantly, to the

ulcer.(97) In CLI, where it is common the involvement of the tibial vessels, ideally, the normal

Figure 18. Angiosome skin-surface representation, courtesy of Desmond Bell in Podiatry

Today.

72

vascular anatomy should be restored with a “complete” revascularization.(166) Nonetheless,

it is not always feasible to achieve.

Thus, in case of being able to choose one vessel, the angiosome theory, first suggested

by Taylor and Palmer,(167) could be helpful. This is particularly important in diabetic and

ESRD patients with poor collateral communication between territories.(168,169) The anterior

tibial and dorsalis pedis arteries supply the tissue comprising the anterodorsal surface of the

calf, ankle, and foot; the posterior tibial artery supplies the plantar surface of the foot and

posteromedial calf, ankle, and heel through its plantar and medial calcaneal branches; and the

peroneal artery supplies the posterolateral calf and ankle as well as the lateral heel (figure 18).

Despite several studies have reported outstanding revascularization outcomes upon

using the angiosome concept in treating diabetic feet,(168,170) there is much controversy

regarding the necessity of angiosome-targeted angioplasty, with multiple studies falling on

both sides of the debate. A meta-analysis by Chae et al.(171) identified four cohort studies

reporting on 881 limbs, comparing the effect of angiosome-targeted angioplasty and indirect

revascularization to treat CLI in diabetics. They found that angiosome-targeted angioplasty

improved limb salvage rate (OR = 2.21, p = 0.001) and wound healing rate (OR = 3.3, p <

0.001). Consistently, several studies have favored the angiosome-targeted angioplasty with

better limb salvage rates.(172–175) On the other hand, when adequate collateral vessels were

present, the outcomes of indirect revascularization were similar to the angiosome-targeted

ones,(176) although leaving the uncertainty of what is exactly the “adequate collateral vessel”

presence.

Further, it worth mention that CLI in diabetic patients frequently develop concentric

continuous vascular calcifications that could limit the effectiveness of endovascular

angioplasty(174) and lead to revascularization of non-targeted vessels. In addition to that,

when indirect revascularization is the only way to improve foot revascularization, it needs to

be done.(171) Last but not least, since there is a reported 9%–12% of anatomical variations,

we should consider them when we want to apply the angiosome theory.(177–179)

After choosing the target vessels, the revascularization could be deemed successful

enough based on several sings. Angiographically we can take the patency of previously

occluded vessels, improved vessel caliber and flow velocity, and decreased collateral flow as

an intended result.(97) The presence of wound blush is associated with higher perfusion,

which improves limb salvage rates.(180) In addition, palpable pulses can be considered an

optimal result. Ultrasound imaging can also be performed after the procedure, with assessment

73

for distal runoff patency and blood flow improvement. Restoration of normal limb coloration,

shortening of capillary refill time under 5 seconds can be seen after a successful intervention.

(97)

1.5. 2D Perfusion angiography

The concept of perfusion is derived from the Latin “perfundere”, composed of “per”

(around) and “fundere” (to pour). In physiology, perfusion refers to the circulation of blood

through the vascular bed formed by biological tissue structures.

2D perfusion angiography (PA) is a modification of an original image processing

software firstly developed to assess the penumbra pattern in acute stroke patients. However,

its application in the CVD arena has been limited because in the brain a local area of ischemia

is surrounded by normal brain, and 2D summation of damaged and normal perfused cerebral

tissue renders precise quantification impossible.(181)

This thesis is focused on the foot dedicated PA software, developed by Philips medical

(Best, The Netherlands) and is an image processing algorithm for DSA images based on the

change of contrast density per pixel over time.(182) The purpose of this tool is to assist the

physician in the diagnosis and quantification of the perfusion alterations in a given patient by

providing a color-coded representation of a DSA run and depicting a curve from the average

of time-density curves for each pixel into a selected Region Of Interest (ROI).

Figure 19. Color mapping of perfusion angiography. The image on the right side

shows the PA before EVT, with the ROI placed over the ulcer-zone; while the

image on the left corresponds to the perfusion angiography run after EVT.

74

Thus, the PA analysis will be visualized by the tool either as a pixel-by-pixel color

map (figure 19), or a time-density graph (figure 20). The tool, therefore, enables an evaluation

of the flow in the imaged tissue but also a comparison between subsequent runs during the

intervention.

From that PA analysis we can obtain 6 parameters, which are numeric measures of the

average time-density curve, and are defined as follows:

1. Arrival Time (AT): AT is the time elapsed between the first frame selected for

processing and the frame where contrast is detected. This is the time between main arterial

inflow and inflow into the small vessels of interest.

2. Peak Time (PT): PT is the time between the contrast uptake point (AT) and the time

at which the contrast has maximum density.

3. Wash-in Rate (WR): The WR is defined by the slope of the uptake curve. The WR

gives an indication of the flow rate or speed in the big foot vessels.

4. Width (W): The W is measured between the inflection points on the wash-in curve

and wash-out curve. The W parameter is an alternative to the Mean Transit Time (MTT)

parameter, but it does not consider any asymmetry in the time-density curve.

5. Area Under the Curve (AUC): The AUC is the area under the curve between the AT

and the point that the contrast has left the vessel.

6. MTT: MTT is the time between AT and the point of the center of gravity in the time-

density curve. This functional parameter is an indication of the average time it takes for

contrast to pass through the tissue.

75

Figure 20. Representation of the PA parameters over the curve: (1) Arrival Time, (2)

Peak Time, (3) Wash-in Speed, (4) Width, (5) Area Under the Curve and (6) Mean Transit

Time.

The foot should ideally be immobilized before beginning the procedure to avoid

extensive movement that will cause the run to be unquantifiable. Methods to ensure complete

immobilization include the use of a dedicated footrest, general anesthesia, or various creative

alternatives with tape, X-ray compatible rest devices, and non-abrasive wraps. Toe movement

can also be excluded when defining the ROI.(181)

Techniques of PA measuring are already used, but we don’t understand the

physiological meaning of the parameters analyzed and we cannot link them up to the type or

degree of stenotic and occlusive lesions and neither to the clinical prognosis. Therefore, PA

cannot be used in decision-making.(183)

77

2. JUSTIFICATION

79

Nowadays there is no objective measure to assess the success of EVT in CLI

patients. PA is an image-processing software which may help in quantifying the degree of

post-EVT perfusion. Thus, our aim is to define an objective goal based on PA to ensure

successful EVT of CLI patients.

Furthermore, a simpler anatomic classification system, able to describe the arterial

disease burden below the groin, needs to be created to assess optimal therapy for any given

patient.

81

3. HYPOTHESIS AND OBJECTIVES

83

HYPOTHESIS

Perfusion angiography can give us objective information on the degree of distal blood

irrigation to the lower limbs in patients with peripheral arterial disease, which is pivotal in

clinical outcomes. This degree of irrigation could be deduced from the perfusion parameters,

which might be also related to patterns of stenotic and occlusive lesions.

It is the hypothesis of this thesis that the information generated in perfusion

angiography could have a higher prognostic value on the healing of ischemic ulcers and

patency of revascularized limbs.

OBJECTIVES

Given the previously described state of the art, the main objectives of this thesis are:

a) to give a clinical meaning to imaging parameters of the perfusion angiography;

and,

b) to set an objective measure to determine success in endovascular therapies;

and,

Following, the secondary objectives of this thesis are:

c) to describe a novel classification with perfusion, clinical and anatomical

information of the arterial steno-occlusions burden in the whole limb; and,

d) to correlate the perfusion angiography parameters with non-invasive

diagnostic methods (mainly the TCPO2); and,

e) to explore the effects of an incomplete plantar arch on perfusion angiography.

85

3. METHODS

87

4.1. Study design

4.1.1. Setting

A prospective cohort study was developed in the Interventional Radiology Department

as a part of a diabetic foot clinic in the Policlinico Abano Terme (Abano Terme, Veneto, Italy).

It was anticipated a study population with high prevalence of diabetic patients with CLI and

an estimated volume of 800 patients per year. The clinical follow-up was at least 6 months

after intervention.

4.1.2. Patient selection

The criteria to select a patient to perform the Perfusion Angiography (PA) analysis

was chosen randomly as only 1 case every 2 consecutive patients (independently from clinical

stage or expected technical complexity), due to daily time schedules limitations for the clinical

practice.

4.1.3. Inclusion criteria

Every patient who received EVT for CLI (with Rutherford classification ranging from

4 to 6) and for very short distance claudication (considered in Rutherford 3) in which the PA

before and after treatment had been performed; with the technical requirements described

above and with comparable initial and final projections.

4.1.4. Exclusion criteria

Firstly, PA studies with artifact because of movements of the foot were firstly

excluded from analysis either because this affects the curve of the PA graph or because the

PA protocol could not be properly done. Further considerations to this point are developed

below in “Perfusion angiography measurement”.

Patients with borderline end-stage renal disease or allergy to iodinated contrast media

were studied with CO2 angiography. Subsequently, those ones were not considered to be

selected.

Those patients without a PA after treatment (not done because of lack of collaboration

from the patient or because a non-standardized PA technique in the after-treatment PA) were

only considered for the hemodynamic analysis, and not for clinical follow-up endpoints.

88

To end with, patients whose pre-treatment PA analysis could not be analyzed were

analyzed for post-treatment PA variables.

4.1.5. Interventions

Interventions were performed under local anesthesia. A percutaneous antegrade

approach of the common femoral artery was taken to place an 11 cm 6F sheath (St. Jude

Medical®, Saint Paul, Minnesota). Standard DSA projections (Philips Allura Xper FD20,

Philips Healthcare®, Best, the Netherlands) will be done pre- and post-EVT. The protocol for

contrast injection consisted in 9 mL of Visipaque 270 mg/ml (iodixanol, GE Healthcare Inc.®,

Princeton, NJ) administered at a speed of 3 ml/s using a coupled injector. The projection was

lateral or anterior-posterior views, with a maximum magnification while minimizing the

distance between the foot and the detector and ensuring the foot remains completely visible.

The projection setting was retained for all PA runs in the same patient.

The injection for the PA analysis was done from the short sheath in the proximal

superficial femoral artery (SFA) to avoid losing the hemodynamic information of the

femoropopliteal (FP) lesions. Ideally, we might have to measure up any eventual aorto-iliac

stenosis and cardiac ejection fraction; but the information obtained was at least as

representative as possible of the hemodynamic of the whole limb. This way, the information

obtained from the study was applicable to decision making on the clinical practice since it’s

the normal scenario in the EVT with intention to treat CLI patients.

The foot of the patient had not been immobilized if there was no need and no

movement artifacts were detected in the PA run. In case of foot movement during de PA, it

will be tied up to the operating table with a tape over a protection to avoid abrasive lesions on

the delicate skin of a CLI foot.

The PA was measured prior to the revascularization and afterwards. PA images were

obtained at a frame rate of 3/s until the operator extends the acquisition. Time of acquisition

depended on a subjective impression that the contrast had adequately arrived to the foot. The

maximum time of acquisition supported for analysis is 30 seconds (90 frames). The image

was analyzed by the Philips software Interventional Workspot 1.3.1 / 2D Perfusion 1.1.6©.

4.1.6. Variables

The collected variables were:

89

- Demographic: Age (years), gender, smoking habits (no/former/active), weight (Kg),

height (cm), Body Mass Index (BMI).

- Comorbidities: diabetes mellitus, hypertension, atrial fibrillation (without pacemaker

rhythm), ischemic cardiomyopathy, chronic kidney disease, end-stage renal disease,

cerebrovascular disease, autoimmune disease.

- Medications: Antiplatelet treatment (no/simple/double), anticoagulation and corticoid

therapies.

- Vascular history: ipsilateral or contralateral endovascular treatment, ipsilateral or

contralateral surgical treatment, contralateral major amputation, CABG and PCI. Also the

baseline Rutherford classification(87) and the WIfI clinical stage(1) for benefit of

revascularization, TCPO2 (mmHg) and Texas Ulcer Classification (TUC)(88).

- Laboratory: Creatinine before and after EVT (mg/dL), Reactive C Protein before EVT

(mg/dL), Hb before and after EVT (g/L)), leucocytes before and after EVT (x109/L) and

neutrophils before and after EVT (x109/L).

- Intervention variables: Plain angioplasty, drug coated balloon, bare metal stent, drug

eluting stent or atherectomy, both in above or below the knee.

- Conventional angiography classifications: TASC classification(15,89) femoral

popliteal (FP) and Below-The-Knee (BTK), pre and post, and the “Abano Terme Score” (ATS)

for each artery of the limb.

- Perfusion Angiography (PA) measurement: Each parameter pre-treatment and post-

treatment, measured both for the whole foot and for the ROI of the ulcer. Quality of the PA

will be also collected.

4.1.7. Clinical follow-up

There were 2 follow-up checkpoints: at 1 and 6 months: Date of visit, TCPO2 (mmHg),

target lesion revascularization (TLR), ulcer healing and date of event, WIfI stage(1) and TUC

classifications(88), amputation free survival, Major Adverse Limb Event (MALE), Major

Adverse Cardiovascular Event (MACE) as a stroke, clinical significant infarction or arterial

thrombosis in any vascular region, Death.

Patients who missed the clinical follow-up of the diabetic foot clinic had been phone

called twice during the data collection in order to find out if they presented eventually a

MALE or they died (not being reported on clinical follow-up). Patients who died during

90

follow-up and those who missed a follow-up control before their ulcers healed, were also

excluded from analysis. If the date in which the ulcer healed could not be inquired but we

know instead that it healed and the limb was salvaged, the case was excluded from analysis

of the time to heal, but included in the general limb salvage rates.

4.1.8. Data collection

Data were collected by the main investigator using a form database software (Ninox

Software GmbH®, Berlin Germany), and several control variables were collected only to

ensure the precision and truthfulness of the collected data. Those variables were data

collection check, marks to review cases or data and marks on the image quality. Data from

one case could not be accidentally merged with another case because of the form entry data

process.

4.1.9. Ethics

Data was collected at that single center and the study was conducted in full compliance

with the Declaration of Helsinki, after being approved by the local Institutional Review Board.

All patients provided informed consent before the procedure.

4.2. Perfusion angiography measurement

4.2.1. Projection

The projection of the PA was mainly done in an antero-posterior view of the foot, but

it could had been also done in a standard lateral projection. In both ways the area had been

the same, including from peroneal bifurcation (at the level of the malleolus) until the proximal

part of the toes. There was no advantage in measuring above the malleolar area because it

would had given information about the flow in a concrete tibial artery, but not about the foot.

Anyway, the antero-posterior projection was preferable to evaluate the terminal blood

flow of the food and to discriminate between precise zones of the ulcers. An exception for

this would have been the case of heel ulcers or in a prior Chopard or Lisfranc amputation, in

which a lateral view for the PA may be better. However, in both AP and lateral projections

the area of the ROI had been theoretically the same, including from peroneal bifurcation (at

the level of the malleolus) until the proximal part of the toes; in order to avoid confounding

information about the flow in a concrete tibial artery, and not in the ulcer or the entire foot.

91

4.2.2. ROI tracing

In case that our aim was to analyze the perfusion to the whole foot, the ROI included

the whole foot arch, from the malleolar level to the toes (excluding outline borders and areas

with movement artifacts). For the perfusion of the ulcer zone, instead, we traced the ROI over

the tissue affected by the ulcer. Subsequently, the angiosome concept had been neglected

because we were measuring the perfusion in the concrete ulcer tissue, regardless of the

anatomical variations and the angiosome targeted revascularization.

The outlining of the ROI had been traced respecting anatomical symmetry of the

arteries between both projections (before and after EVT). If we excluded a concrete zone of

the foot from the ROI to avoid movement artifact, the ROI in the comparative PA image also

excluded the same zone, even if there is no artifact in the latter. This way both images were

comparable from the hemodynamic point of view.

4.2.3. Time of exposure

Regarding the time of the exposure and analysis of the PA, it’s important to consider

that this time was decided from the arbitrary duration of the Digital Subtraction Angiography

(DSA) run and it depended on the contrast arrival. This duration was a decision took during

the intervention and cannot had been equalized for all patients. Thus, the more adjusted

measurement to the limb’s real state was to analyze the whole duration of the PA run.

The color bar scale had been the same in both PA images and fully ranged in order to

improve the resolution. Nevertheless, it did not modify the value of the parameters that are

the object of study. It does not seem to be useful to set trim the exposure time too much

because the behavior of the PA graph was not coherent and might be a consequence of some

gap into the software algorithm.

To avoid artifacts, time of exposure was cut-off from the ending tail, but not from the

start because in the latter fashion time dependent parameters would have been wrongly

modified.

In the cases with runs longer than 90 frames, only the last 30 seconds were analyzed

and the delay time before analysis should was added to the arrival time value. The other

affected parameter by this hypothetical issue was the Area Under the Curve (AUC), but it

could had not been corrected because we can add the time, but the AUC in the deleted frames

of long series were not recorded. The rest of the parameters, since they only depend

92

theoretically on the morphology of the curve they would remain invariable after a time delay

before analysis.

4.2.4. Artifact management

The PA image will be considered optimal when we can only see the arteries in the PA

image. That implies that the bone or the skin could not be seen and the curve follows a

physiological morphology.

It was considered as an artifact any mark of the bone’s or skin’s outlines in the PA

image, also the speckled color tracing in the zone of the bone marrow. Anyway, the magnitude

of those artifacts was dependent on how much is the time-density curve modified. Here, the

artifact was seen as a peak appearing through the curve. Usually they could be seen isolated

but if there were too many, the peaks could not be identified, thus, the curve was useless and

should be excluded. That is, the artifact that revealed a unique movement was reflected on the

curve as a unique peak; on the other hand, the artifact that depicted a speckled color tracing

in the PA image was represented as multiple small peaks that could have increased the slope

of the curve and might have modified the values of the parameters.

If the maximum density in the curve did not match in time with the corresponding

color (codified by the time-color bar) with which the foot arteries were depicted, it would

have mean that the curve was not representative of the arterial flow but modified by an artifact.

Subsequently, the run was not valid and had been excluded. The other face of the coin of this

fact was used as a checkpoint, ensuring that the color corresponding to the time of the

maximum point in the curve was matching the main color of the PA image. Another strategy

used to test the truthfulness of the time-density curve was to scroll the time of exposition

(from second 0.3 till the end of the run) and observing that the arteries appeared in the PA

image synchronized with the graph curve, denoting that the curve reflected the contrast in the

arteries and it had no relation with any other artifact.

If the movement artifact was clearly limited to a zone in the foot, the ROI had been

drawn excluding that zone and the PA took to analysis. Symmetrical ROIs were always taken

from the PA image before and after treatment. To ensure this symmetry, we took anatomical

references (arterial bifurcations, shape, etc.) and not with the provided chained tool, in order

to increase precision.

93

4.3. Conventional angiography: measurement and classification

The standard DSA projections of the limb were those described by Manzi et al.(184)

The angiographies were recorded as a maximum opacity pictures and stored for later

classification according to TASC and the below described score.

4.3.1. TASC classification

The angiographies before EVT and thereafter were classified according the TASC

classification(89) for both FP and BTK sectors. The TASC classifications does not have any

grade to describe a patient with no stenosis; therefore, it was classified as a “no lesions

angiography”, besides the TASC categories A, B, C or D. The criteria for each grade of the

classification in the FP sector were:

- TASC A: single stenosis ≤ 10 cm or occlusion ≤ 5 cm in length

- TASC B: multiple lesions (stenosis or occlusions) ≤ 5 cm each, single lesion ≤ 15

cm (not involving infra-geniculate popliteal artery) or heavily calcified occlusion ≤ 5 cm.

- TASC C: multiple lesions totaling ≥ 15 cm

- TASC D: chronic total occlusions of SFA ≥ 20 cm or popliteal artery and proximal

trifurcation vessels.

A specific consideration on femoral popliteal TASC B classification was that a short

stenosis in the popliteal artery will be considered as TASC A, in order to give it a more

hemodynamic sense, rather than a technical one. In TASC D, the possibility of an occlusion

of the common femoral artery (CFA) is not considered because the EVT is not suitable on the

CFA occlusions.

The TASC classification for the below-the-knee sector obeyed the following criteria

over the target vessel (assuming worse stenosis or CTO in the other arteries):

- TASC A: stenosis ≤ 5 cm in length

- TASC B: multiple stenosis of ≤ 5 cm each with a total length ≤ 10 cm or a single

occlusion ≤ 3 cm in length

- TASC C: multiple stenosis or a single occlusion of > 10 cm in length

- TASC D: the same lesions of TASC C but with dense lesion calcification or no

visualization of collaterals

94

To study the differences between pre and post TASC classifications, categorical

values of 0, 1, 2, 3 and 4 were given to “No lesions”, “A”, “B”, “C” and “D”, respectively.

4.3.2. Hemodynamic angiography based “Abano Terme Score”

We described a score to ease the classification and to better describe the lesions of the

whole limb, as well as to include hemodynamic parameters.

It was inspired on the Joint Vascular Societies Council (JVSC) classification described

by Toursarkissian et al.(185) referring to the lesions on the out-flow arteries in the foot. The

JVSC classification gave a score from 0 to 3 to each artery on the foot (dorsalis pedis from

the ankle joint level to the level at which it gives off its arcuate branch, lateral plantar, and

medial plantar) and the calf (anterior tibial, posterior tibial and peroneal including the

tibioperoneal trunk) on the basis of the most severe stenosis depending on the following

grading:

- 0 points in a stenosis of 0-20%

- 1 points in a stenosis of 20-50% or in tandem stenosis of <20%. This value will be

also given to no flow-limiting dissections or mild recoiling after angioplasty.

- 2 points in a stenosis of >50% or in tandem stenosis of <50%

- 2.5 points in an occlusion of <50% of the vessel length

- 3 points in an occlusion of >50% of the vessel length

Afterwards, a “foot score” (sum of the scores obtained for the three foot vessels plus

1) and a “calf score” (sum of the scores obtained for the three calf vessels plus 1) were finally

obtained in the JVSC system. Our attempt was to adapt this score to the whole limb. Based

on the same grading criteria, we provided a punctuation to each artery from the groin to the

ankle (superficial femoral artery, popliteal, anterior tibial, posterior tibial and peroneal

arteries). For convenience in this manuscript, we named it as “Abano Terme Score (ATS)”.

The total scoring for the whole limb was obtained by adding up the score given to the

FP and the BTK region. To obtain the score of the FP region we added the squares of the

scores of the SFA and the popliteal artery. In the BTK area instead, we multiplied the scores

of the anterior and posterior tibial arteries, and then the score for the peroneal artery was added.

The rationale behind this strategy is that the presence of lesions in the FP area is

hemodynamically more deleterious, while the lesions in one tibial artery could be properly

compensated by the other tibial artery if the latter has no stenosis. The peroneal artery alone,

95

otherwise, could not replace both tibial arteries. The formula to calculate the global limb ATS

score is explained in figure 21.

According to this scoring, in the case of a long occlusion of all the anterior tibial

(deserving a score punctuation of 3), a posterior tibial without any stenosis and good outflow

(corresponding to a punctuation of 0) and an isolated stenosis of >50% in the peroneal artery

(punctuation of 2), the score for BTK in this case would be 2 (since 3×0+2 = 2). Some

examples are shown in figure 22.

In case of anatomical variations, typically the hypoplasia of the anterior or the

posterior tibial arteries, the score value is given to the artery that procures blood supply to the

theoretical angiosome of the hypoplastic artery. The latter is usually the peroneal artery. The

tibio-peroneal trunk will be scored as a part of the posterior tibial artery.

Prior FP bypasses had been scored as if they were the native FP, with eventual stenosis

or occlusions, provided that they are equivalent from the hemodynamic point of view.

The advantages to evaluating limbs using this type of score are several. First, an index

of the global hemodynamic significance of the lesions will be obtained, aiding in the

correlation with the degree of perfusion. Indeed, it is mandatory to correlate hemodynamics

and perfusion to accurately quantify the severity of the lesions.

Second, the score provides a simplified report of the steno-occlusion pattern in a limb,

which is very useful in daily clinical practice and in the standardization of descriptive analysis

of the disease burden of a limb. For instance, if we want to describe an IC patient we can

understand that a compatible score would be a 2.5+1+0+0+0; or, on the other side of the coin,

a typical CLI pattern in a diabetic patient would be 2+0+3+3+0.

Figure 21. Formula for the global limb ATS. ATS: Abano Terme Score; ATA: anterior

tibial artery ATS; ATP: posterior tibial artery ATS; Per: peroneal artery ATS; Pop: popliteal

artery ATS; SFA: superficial femoral artery ATS.

𝑙𝑖𝑚𝑏𝐴𝑇𝑆 = 𝑆𝐹𝐴+ + 𝑃𝑜𝑝+ + (𝐴𝑇𝐴 × 𝐴𝑇𝑃+ 𝑃𝑒𝑟)

96

4.4. Endpoints and groups for analysis

The independent variables of the study were the PA parameters after the EVT. They

were measured on the whole foot and in the ROI corresponding to the ulcer. Thus, the

angiosome concept was not applied since the area of tissue measured was straightly the ulcer,

independently to the tibial vessel to be treated. To avoid confounding factors, such as central

hemodynamic variables, we also annotated the delta (∆) changes (increases or decreases) of

the values of a certain PA parameter.

TASC classification and ATS values were annotated separately and with the changes

through the EVT (∆).

Figure 22. Examples of the ATS system. Composition of images from DSA conventional angiographies. The projections vary from the

above the knee and the ones below the knee. The ATS is calculated in the two cases on the left to show its descriptive ability regarding

the disease burden in a limb, being (A) 0+0+0+3+3 (ATS for the whole limb=3) and (B) 3+3+0+3+3 (ATS for the whole limb=21). The

two cases on the right demonstrate the ATS’ capacity to describe the global post-procedural improvement of the limb. In the case (C),

the ATS score improved from 13 (2+3+0+3+0) to 0 (0+0+0+0+0). The case (D) improved its ATS of 28 (3+3+3+2.5+2.5) to 3

(0+0+3+0+3). ATS: Abano Terme Scores; DSA: Digital Subtraction Angiography.

97

The dependent variable for the main endpoint was the presence of an ulcer or not. In

cases without an ulcer, as in Rutherford 3 and 4 patients, the groups were divided if they

needed a target lesion revascularization (TLR) or not. Herein, the endpoint is not referred to

the concrete lesion but a reintervention on the same limb.

On the other hand, in patients with and ulcer, the dependent variable was the time to

heal (TTH), and patients were classified according to whether the TTH is lower (group A) or

higher (group B) than a specific amount of days. The optimal TTH interval was explored from

several time cut-off points: 30 days, 40 days and 120 days. The reasons for choosing several

cut-off points was the uncertainty on the prediction ability of PA, besides from other variables

that contribute to the healing process (cures, nutrition status, etc). The 30- and 40-days cut-

off points were chosen due to their easiness to determine precisely the healing process because

of more frequent controls in the first weeks. The 120 days cut-off was chosen as suggested by

Reed et al.(186) who found a higher probability of MALE in patients who did not heal by 4

months. As a last consideration, if a patient has required a TLR or a major amputation, the

case was be considered as non-healing, thus, included in group B.

4.5 Statistical analysis

Descriptive and frequency statistical analysis were obtained and comparisons were

made by use of the software IBM SPSS Statistics 22.0®. Categorical variables were reported

as frequencies (percentages) and continuous variables as mean ± SD or median (interquartile

range), as appropriate. Distributions of the PA parameters were checked for normality using

Q-Q plots. Box-Plots were used to represent values of quantitative variables among the study

population.

Changes in TASC classification were assessed with the McNemar-Bowker´s test. Pre-

post changes of PA parameters (before and after EVT) were assessed by the paired samples

t-test. Intergroup differences between dependent variables (TTH) and PA parameters were

performed using the Student’s t test. Receiver characteristic operator curves (ROC) were

configured in order to calculate cut-off points for PA parameters with best sensitivity and

specificity to predict a best outcome. These variables were then categorized and compared to

TTH with the Pearson’s chi-square or the Fisher’s exact test, as appropriate. Kaplan-Meier

curves were performed to evaluate the time of ulcer healing depending on the different values

of PA parameters, determining the log-rank test to assess statistical significance between

groups.

98

Logistic regression models adjusted by potential confounding factors were performed

in order to identify variables which could be independent predictors of TTH. Predicted

probabilities of the model were obtained and a ROC curve was configured in order to measure

its predictive ability.

Associations between TCPO2 and values of PA and ATS parameters were assessed

with Spearman’s correlation coefficients.

Pre-post changes in Classification Scores (ATS) were evaluated with the Wilcoxon

test. ATS were compared with TTH using the Mann-Whitney U test. Spearman’s correlation

coefficients were used to assess correlations between classification Scores (ATS) and PA

parameters; for this purpose, a specific database with both pre and post values evaluated as

different cases was created.

A p-value<0.05 was considered as statistically significant.

99

5. RESULTS

101

5.1. Description of the population

From January 2015 to July 2016, out of 1189 patients treated for CLI, 580 patients

were selected for a PA study protocol. The complete PA protocol described for this project

was achieved in 332 patients; but not in the rest due to several reasons: popliteal selective

injection, hand injection (without using the coupled injector), non-standardized projections,

lack of preoperative or postoperative runs or significant differences between them on the

projection or excessive movement artifact detectable before performing the PA analysis.

Among cases analyzed with PA, 39 patients were excluded in a second step: 21

patients died with unhealed ulcers and 18 were lost during follow-up. Among the missing

cases we suspect three possible reasons: the patient had been managed in another hospital or

city (with clinical success or requiring an amputation) or had died, although the latter could

not be confirmed through the final data review phone call).

Figure 23 shows an algorithm of the aforementioned exclusions before acquiring the

baseline study population (N=293) for specific analysis. Among this population we found 250

cases with ulcers (Rutherford 5 and 6 classifications), where the main endpoint used was the

TTH; and 44 other cases without ulcers (Rutherford 3 and 4), compared to the TLRs as the

main endpoint.

Figure 23. Algorithm of exclusions, from every patient with a PA run and undergoing an EVT to cases fit to

be analyzed. CLI: Critical Limb Ischemia; EVT: Endovascular Treatment; PA: Perfusion Angiography.

102

In the cases where only the post-treatment PA parameters were used for analysis, only

one correct post-treatment image was required, obtaining an N=246. On the other hand, when

the analysis was to be over the ∆ PA parameters, a correct pre-treatment image was also

required, thus, the final population was N=222. In each case, as it is shown in figure 24, cases

were divided again whether they had ulcers or not.

Figure 24. Subgroups for analysis of the perfusion angiography. PA: Perfusion Angiography.

In the baseline study population (N=293), the mean age was 72 years and 67.5% were

men. Common comorbidities were diabetes mellitus (DM), hypertension, chronic kidney

disease (CKD), end-stage renal disease (ESRD), atrial fibrillation (AF), ischemic

cardiomyopathy (ICM), cerebrovascular disease (CVD) and autoimmune diseases (AD).

Missing data regarding smoking habits was lacking (information was retrieved in only 20.2%

of the cases) and thus the calculated prevalence was assumed not representative. Data on any

previous revascularizations (of the limb or the coronary arteries) and minor ipsilateral

amputations or major contralateral amputations was also collected. The frequencies of all of

them, as well as the antiplatelet, anticoagulation and corticoid therapies, are described in Table

1.

103

Table 3. Baseline characteristics of cases for analysis. Data are

shown as N(%) or mean±SD. AF: Atrial Fibrillation; BMI: body

mass index; CABG: coronary artery bypass graft; CKD: Chronic

Kidney Disease; CLI: Critical Limb Ischemia; CVD: Cerebro-

Vascular Disease; DM: Diabetes Mellitus; ESRD: End-Stage Renal

Disease; ICM: Ischemic CardioMyopathy; PCI: percutaneous

coronary intervention.

Table 4. Laboratory tests did not reveal any clinically relevant

difference due to the EVT. Data are shown as mean ±SD.

Table 5. EVT variables. Data are shown as N(%) or mean±SD.

BMS: bare metal stent; BTK: Below-the-knee; DCB: drug coated

balloon; DES: drug eluting stent; POBA: plain old balloon

angioplasty.

Table 6. Baseline clinical stages frequencies. Data are shown as

N(%). TCPO2: TransCutaneous Pressure of Oxygen; TUC: Texas

Ulcer Classification; WIfI: Wound Ischemia foot Infection.

104

Laboratory tests and the kind of EVT performed are shown in Tables 2 and 3,

respectively. At the baseline clinical stage, Rutherford classification(87), the Texas Ulcer

Classification (TUC) (88) and the WIfI clinical stage(1) were annotated and described in

Table 4. The plantar arch had been cropped in 11 cases (3.8%) and we found 52 cases (17.7%)

with anatomical variations. Proximal arterial occlusive disease was detected in 14 cases (4.8%)

and a there was a pre-planned amputation in 170 cases (58%).

5.2. Distribution of the endpoints and PA parameters

As the main endpoint, for the cases with ulcers, we have chosen the TTH and its mean

was 75.9 ± 85.9 days and the process of healing of ulcers is described in figure 25. The

equivalent endpoint in Rutherford 3 and 4 cases was freedom from TLR, which was observed

in 23 out of 37 (62.2%) patients during follow-up months.

The distribution of TASC classifications, ATS scores and WIfI clinical stages(1) are

represented in figures 26, 27 and 28, respectively. The subgroup used here was that without

exclusion criteria and with a good quality image of PA run after the EVT (N=246).

Figure 25. Ulcer healing rate. One minus survival curve for the healing of the ulcers during follow-up.

The analyzed data is over the ulcer healing rate at 6 months, which was the 70.6%. TTH: Time to Heal.

105

Figure 26. TASC classifications of the before and after treatment angiographies, in the

femoropopliteal (FP) and in the below the knee (BTK) regions. EVT: Endovascular Treatment.

Figure 27. Evolution of WIfI clinical stage preoperatively and during follow-up. WIfI: Wound

Ischemia Foot Infection.

106

Figure 28. Box-plots of ATSs, either for pre- and post-EVT. ATS: Abano Terme Score; BTK: Below-The-Knee; EVT: Endovascular

Treatment; FP: Femoro-popliteal.

Table 7. Changes in TASC, ATS and PA parameters from before to after EVT. Data are shown as mean ± SD. AT: Arrival Time; ATS:

Abano Terme Score; AUC: Area Under the Curve; BTK: Below-The-Knee; FP: Femoro-popliteal; MTT: Mean Transit Time; PA: Perfusion

Angiography; PT: Peak Time; W: Width; WS: Wash Speed.

107

The means ± SD of TASC, ATS and PA parameters measured on the foot and at the

ulcer are all exposed in Table 7. All the parameters presented statistically meaningful

differences through the EVT. The frequencies of PA parameters after EVT in the study

population are represented in figure 29a for those parameters measured in seconds, and in

figure 29b for those other parameters which have no units. For this picture the group used was

also the N=246; as in the previously mentioned classifications. All values measured on the

ulcer, except the PT, showed significant differences, meaning that they are measures affected

by the EVT revascularization. In Table 6, instead, we can see the changes in ATS and PA

parameters through the EVT.

Figure 29a. Box-plots of PA parameters with seconds as units of measure, either pre- and post-EVT runs. AT: Arrival Time; EVT:

Endovascular Treatment; MTT: Mean Transit Time; PT: Peak Time; W: Width.

Figure 29b. Box-plots of PA parameters witout units, either pre- and post-EVT runs. AUC: Area Under the Curve; EVT:

Endovascular Treatment; WS: Wash Speed.

108

Table 8. Distribution of ∆ATS and ∆PA para-meters. Data are shown as mean±SD. AT: Arrival Time; ATS: Abano Terme Score; AUC: Area Un-der the Curve; BTK: Below-The-Knee; FP: Fe-moro-popliteal; MTT: Mean Transit Time; PA: Perfusion Angiography; PT: Peak Time; W: Width; WS: Wash Speed.

Table 9. Follow-up events frequencies. The N=332 was used for missings and deaths. Data are shown as N(%) or mean±SD. AFS: Amputation Free Survival; MACE: Major Adverse Cardiovascular Event; MALE: Major Adverse Limb Event; TCPO2: TransCutaneous Pressure of Oxygen; TLR: Target Lesion Revascularization.

5.3. Follow-up

The follow-up events frequencies and means are explained in Table 9. Only the final

study MALE rate was collected. A total of 188 (64%) of minor amputations were performed.

Due to the low rate of major amputations (3.4%), a Kaplan-Meier curve for the amputation

free survival (AFS) is not represented. Among the 21 cases that died, the mean days to the

event were 177 ± 167 days.

109

5.4. Predictive ability of clinical success in patients with rest pain

There were 37 patients with Rutherford 3 and 4 classifications with a good image in

the PA post-EVT, the baseline characteristics for this group are exposed in Table 8. Among

them we found 11 cases (29.7%) of anatomical variations, while the proportion of anatomical

variations in cases with ulcers (17.7%) was lower, although there were no statistical

differences (p=0.146). Proximal arterial occlusive disease was present in 2 cases (5.4%). No

cases had a previously surgical cropped plantar arch.

Table 10. Baseline characteristics of the subgroup of patients

without ulcers (Rutherford 3 and 4). Data are shown as N(%) or

mean±SD. AF: Atrial Fibrillation; BMI: body mass index; CABG:

coronary artery bypass graft; CKD: Chronic Kidney Disease; CLI:

Critical Limb Ischemia; CVD: Cerebro-Vascular Disease; DM:

Diabetes Mellitus; ESRD: End-Stage Renal Disease; ICM: Ischemic

CardioMyopathy; PCI: percutaneous coronary intervention.

Table 11. Differences in post-EVT and ∆PA parameters for TLR during follow-up. Data are shown as N(%) or mean±SD. Patients with good post-EVT image were N=37 and patients with good both pre- and post-EVT were N=32. AT, PT, W and MTT are expressed in seconds. AT: Arrival Time; AUC: Area Under the Curve; MTT: Mean Transit Time; PA: Perfusion Angiography; PT: Peak Time; TLR: Target Lesion Revascularization; W: Width; WS: Wash Speed.

Table 12. Comorbidities and possible confounding factors among cases without ulcers and suitable for PA analysis (N=37). The percentages ar in relation with the column N. AF: Atrial Fibrillation; CKD: Chronic Kidney Disease; CVD: Ce-rebro-Vascular Disease; DM: Diabetes Mellitus; ESRD: End-Stage Renal Disease; ICM: Ischemic CardioMyopathy.

110

Comorbidities were equally distributed between cases which required a TLR and those

that did not, as it is shown in Table 9. Anyway, similar results (exposed in Table 10) came out

from exploring the predictive power of PA parameters for requiring a TLR.

5.5 Predictive ability of clinical success in patients with ulcers

The TTH used to divide groups A and B was first calculated at 30 days. Then, when

using the post-EVT PA parameters, only the cases with an adequate post EVT image of PA

were used (N=209). The baseline characteristics, Rutherford(87) classifications and WIfI

clinical stages(1) showed no significant differences between groups (table 11). In a

subsequent analysis we took cases with adequate images both before and after EVT (N=190)

to use then the increments in the parameters (∆).

The results of the comparisons between TTH and post-EVT ∆PA parameters are

shown in Table 12, where we found significant differences in AT, PT, MTT, ∆PT, ∆W and

∆MTT.

Table 13. Comorbidities and possible confounding factors. Among cases with ulcers and suitable for PA analysis (N=209). WIfI High was compared to non-High and the Low and Moderate ones with other stages particularly. AF: Atrial Fibrillation; CKD: Chronic Kidney Disease; CVD: Cerebro-Vascular Disease; DM: Diabetes Mellitus; ESRD: End-Stage Renal Disease; ICM: Ischemic CardioMyopathy; TTH: Time To Heal; WIfI: Wound Ischemia Foot Infection.

Table 14. Differences in post-EVT and ∆PA parameters for TTH at 30 days. Data are shown as mean ± SD. Patients with good post-EVT image (N=209) for post-EVT and good both pre- and post-EVT (N=190). Units of AT, PT, E and MTT are expres-sed in seconds. Mann-Whitney was used for WS, W and for ∆ parameters. AT: Arrival Time; AUC: Area Under the Curve; MTT: Mean Transit Time; PA: Perfusion Angiography; PT: Peak Time; TTH: Time To Heal; W: Width; WS: Wash Speed.

111

Using ROC curves, represented in figure 30, we retrieved the best sensitivity and

specificity cut-off points for post-EVT PA parameters, which are described in Table 13. The

ones that showed statistical differences were analyzed with a chi-square test to calculate the

sensitivity and specificity of each one, also described in Table 13.

Figure 30. Receiver Operating Curves for the post-EVT PA parameters prediction ability for a

TTH<30 days. EVT: Endovascular Treatement; PA: Perfusion Angiography; TTH: Time To Heal.

112

Table 15. Chi Square analysis for cut-off points, which were retrieved from ROC curves.

The differences were also explored for the TTH at 40 and 120 days, as shown in Tables

16 and 17 respectively, without observing statistically significant differences.

Finally, figure 31 shows Kaplan-Meier curves comparing the TTH of ulcers with the

cut-off points of each PA parameter analyzed.

Table 16. Differences in post-EVT and ∆PA parameters for TTH at 40 days. Data are shown as mean±SD. Patients with good post-EVT image (N=209) for post-EVT and good both pre- and post-EVT (N=190). Units of AT, PR W and MTT are expressed in seconds. AT: Arrival Time; AUC: Area Under the Curve; MTT: Mean Transit Time; PA: Perfusion Angiography; PT: Peak Time; TTH: Time To Heal; W: Width; WS: Wash Speed.

Table 17. Differences in post-EVT and ∆PA parameters for TTH at 120 days. Data are shown as mean±SD. Patients with good post-EVT image (N=209) for post-EVT and good both pre- and post-EVT (N=190). Units of AT, PR W and MTT are expressed in seconds. AT: Arrival Time; AUC: Area Under the Curve; MTT: Mean Transit Time; PA: Perfusion Angiography; PT: Peak Time; TTH: Time To Heal; W: Width; WS: Wash Speed.

113

Figure 31. Kaplan-Meier survival curves comparing the TTH of ulcers (days) with the cut-off points of each PA para-meter with statistical differences. AT and ∆MTT appear as the better predictors, according to the logistic regression analysis. AT: Arrival Time; AUC: Area Under the Curve; MTT: Mean Transit Time; PA: Perfusion Angiography; PT: Peak Time; TTH: Time To Heal; W: Width; WS: Wash Speed.

114

5.5.1. Multivariate analysis and confounding factors

The multivariate analysis was made with probable confounding factors such as sex,

age, BMI, DM, hypertension, AF, CKD, ESRD, ICM, CVD, autoimmune disease and WIfI

stage(1) at baseline. The Odds Ratio (OR) values of the statistically significant cut-off points

are shown in Table 18. The cut-off values of the parameters AT>6 s, WS>20, MTT>4.1 s,

∆PT>1.5 s and the ∆MTT>1.7 s maintained the null benefit out of the confidence interval and

persisted with statistically significant differences.

After that step, a logistic regression revealed the ∆MTT of more than 1.7 seconds and

the AT over 6 seconds to be independent predictors of wound healing. The predicted ability

for these two parameters is represented in figure 32, which had an area under the curve of

0.65.

Table 18. Multivariate analysis with possible confounding factors. Used to select the best pre-dictors of wound healing. AT: Arrival Time; AUC: Area Under the Curve; MTT: Mean Transit Time; PT: Peak Time; TLR: Target Lesion Revasculariza-tion; W: Width; WS: Wash Speed.

Figure 32. Multivariate ROC curve of

independent predictor of wound healing:

∆MTT>1.7s and AT>6s; area under the curve =

0.65. AT: Arrival Time; MTT: Mean Transit Time.

115

5.6. Correlations of PA parameters with TASC and ATS

Non-statistical associations were found between ATS and the clinical outcomes,

measured as a TTH of 30 days. Table 19 shows the results from the student T test for ATS

after the EVT and the variation of the ATS.

Table 19. ATS predicting power for wound healing. ATS post-EVT and the increments of ATS are

not related to a higher proportion of ulcer healing in 30 days. ATS: Abano Terme Score; BTK: Below-

the-knee; EVT: Envovascular Treatment; FP: Femoro-popliteal;

On the other hand, a specific database was built with all the cases with PA runs suitable

for analysis (following the described protocol and without artifacts). On this database a N=537

of PA analysis was obtained. In this group, the ATS appeared to be correlated with PA

parameters from the spearman’s rho test, exposed in Table 20. Although there were significant

correlations with TASC classifications, both in FP and BTK regions, the ATS was more

strongly correlated and a unique value could be given to the whole limb (which was the best

correlated value). Concretely the AT and the AUC/s showed a medium-moderate grade of

correlation with the ATS for the whole limb.

Table 20. Spearman’s rho correlation between PA parameters and classification systems: TASC and ATS. *: PT, W and MTT showed a non-significant correlation with ATS of BTK (p>0.05). AT: Arrival Time; ATS: Abano Terme Score; AUC: Area Under the Curve; BTK: Below-the-knee; EVT: Envovascular Treatment; FP: Femoro-popliteal; MTT: Mean Transit Time; PT: Peak Time; W: Width; WS: Wash Speed.

116

5.7. Non-invasive diagnostic methods

The distribution of the TcPO2 is represented in the histogram of figure 33. Note that

due to the retrospective nature of the study only few cases had a TcPO2 measure. The values

of the TcPO2 at baseline, one and six months (expressed in mean ± SD) are 22 ± 15.8 mmHg,

50.9 ± 11.9 mmHg and 32.2 ± 17.2 mmHg, respectively.

From the Spearman’s rho analysis at baseline no significant associations were found.

However, at the 30 days follow-up control of TcPO2 significant correlations were found for

AT after the EVT (sr=0.23, p=0.029), the ATS of BTK (sr=0.25, p=0.020) and for the whole

limb (sr=0.24, p=0.023); all of which are low-medium degrees of correlation.

Figure 33. Histograms of TcPO2 measures at baseline (N=46), at 30 days (N=87) and at 6 months (N=34)

of follow-up. TcPO2: TransCutaneous Pressure of Oxygen.

117

5.8. Effect of incomplete plantar arch on PA parameters

The comparison between the PA parameters in patients with a native pedal-plantar

arch and those in patients with a previous surgically cropped plantar arch (from a minor

amputation) was made with the specific database with both pre- and post-EVT values

evaluated as different cases (N=537). We found 25 cases of angiographies with a cropped

plantar arch.

As it is shown in Table 21, the parameters that were significantly related with an

incomplete plantar arch were those that refer to the slope of the PA curve: the PT (p=0.005),

the WS (p=<0.001), the W (p:0.043) and the MTT (p=0.003).

Table 21. Relation between the PA parameters

and the state of the plantar arch, whether it is

in its native state or it is surgically cropped. Data

are shown as mean ± SD. AT: Arrival Time;

AUC: Area Under the Curve; MTT: Mean Transit

Time; PA: Perfusion Angiography; PT: Peak

Time; W: Width; WS: Wash Speed.

119

6. DISCUSSION

121

The aim of this project is to solve the current challenges that physicians face when

performing an EVT in a CLI patient. These are partially solved by the WIfI clinical staging(1),

which addresses the doubts centered on the outpatient scenario rather than the operating room

uncertainties. Indeed, a diagnostic tool aiding the interventing physician as the WIfI stage

helps the outpatient clinic is still lacking. Therefore, PA parameters may play a central role in

objectively quantifying the degree of distal blood irrigation to the lower limbs, ultimately

simplifying the decision-making process during revascularization based on a threshold value

capable of predicting the healing of an ulcer or the avoidance of amputation. Solving the kno-

wledge gap between revascularization and clinical outcomes may simplify revascularization

strategies and treatment aggressiveness. Furthermore, a numeric and simpler classification

method of the arterial occlusive disease burden in the limb, like the ATS, will allow for com-

parisons with other cases, increase precision in reports and accurately determine the optimal

treatment for each patient. As the WIfI consensus document(1) states, a more precise classi-

fication to describe the disease burden is needed to predict outcomes and allow comparisons

between patients and therapies.

6.1. Results compared with other perfusion angiography studies

Statistically significant changes were found between EVT results of the TASC, ATS

and PA parameters, which supports their validity in the evaluation of distal blood irrigation

of the lower limbs.

While no other risk factor or classification stage at baseline predicted a higher

healing rate before 30 days, with an AT > 6 seconds after the EVT, the ulcer is 2.64 times

more likely to heal in less than 30 days, and if the ∆MTT is > 1.7 seconds, that Odd Ratio is

3.21. The MTT seems to take importance on the difference between before and after EVT,

but if the pre-EVT run of the PA is not adequate, the MTT post-EVT over 4.1 seconds

should be used, being then 3.19 times more prone to heal before a month. The PT and the

∆PT also showed significant differences which further enforces the validity of MTT as they

are measuring almost the same feature on the curve: while the PT measures the time elapsed

between the AT and the maximum peak of contrast intensity in the ROI, the MTT measures

the time elapsed between the AT and the point of the center of gravity in the time-density

curve (see Figure 20 explaining the source of the PA parameters). Otherwise, PA parameters

in the revascularization of rest pain patients did not reveal any significant differences regar-

122

ding clinical outcomes. Not surprisingly, while PA parameters convey information descri-

bing local tissue perfusion at the moment of the revascularization, clinical outcomes of EVT

in patients with rest pain rely on reintervention rate and limb salvage, which involves the

whole hemodynamic status of the limb. However, the observed results in ulcer patients

might be applicable as a composite with other factors to patients with rest pain; an alterna-

tive that deserves investigation in a larger population of rest pain patients.

There are four main studies exploring the feasibility and the clinical significance of

PA, from 2015 to 2017. While three of them explored PA functionalities in the foot as the

target organ of the revascularization, describing the injection technique with a long sheath

placed at the mid-part of the popliteal artery;(181) the forth one used PA software to explore

the effect of certain stenosis measuring only the artery, injecting the contrast medium from

the groin(187). Similarly to the latter, Kostrzewa et al.(188) explored flow-limiting stenosis

with a different parametric color coding software based on angiography (syngo iFlow,

Siemens Healthcare GmbH, Erlangen, Germany). Unfortunately, the differences between the

parameters and the processing principles used render any comparison unfeasible. The

previous studies used the former version of the software (Interventional Workspot R1.1 or

1.0.1) we employed in our project (Interventional Workspot 1.3.1 / 2D Perfusion 1.1.6). This

change in software version impeded the measurement of the peak density value, taken as a

main variable in their studies, but considered prone to movement artifacts and subject to

variability of tissue composition, anatomy and foot positioning.(189) From 9 to 15 cc of non-

ionic iodinated contrast material were injected at a rate of 3 cc/sec (except one in which hand

injections were performed(187)), totaling from 300 to 320 mg I /mL (while in our study the

total dose was of 270 mg I/mL). These differences should not have induced any bias in the

results.

The study population of those studies was of 18(181), 21(187,189) and 89(183)

patients. In none of them a follow-up for clinical outcomes was made. Thus, our study is the

first to evaluate the clinical outcome of EVT in CLI after PA analysis.

With the use of dedicated footrest for immobilization during PA, the rate of invalid

runs due to artifacts was relatively low (ranging 5-14%). The rate of invalid runs in our study,

where we did not use more than fixing the foot with a tape, was slightly higher: 16% when

using only the completion angiography and 24.2% when both pre- and post-EVT runs were

used. However, those exclusion rates did not impede reaching the intended results stated in

the study’s objectives.

Choosing the proper ROI is matter of debate, and while some groups compare the

hindfoot and forefoot regions(189), others suggest that the ROI should include the hindfoot

123

while excluding the forefoot (as the area distal from the mid-metatarsal region) in order to

avoid potential small movement artefacts(181). The technique in itself has been considered to

have limitations to describe local ischemia at the site of an ulcer or to quantify local

improvement after PTA.(181) In spite of all, knowing the three-dimensional impreciseness

when referring to the underlying tissue of an ulcer and assuming more risk to get movement

artifact on the run, we thought that the most representative measure of the selective perfusion

of an ulcer would be measuring over it, independently of its location in the foot. Our

preference is for the antero-posterior or lateral projection depending on the location of the

ulcer. Further, the vast majority of ulcers were located at the toes and on the antero-posterior

projection the amount of tissue overlapping to the target tissue was negligible.

Previous studies have been controversial regarding the best parameter on which to

base the decision-making during the procedure. While Murray et al.(189) only found a

significant increase after successful angioplasty in the AUC (29.4%, p=0.03); Reekers et

al.(183) described an average increment of 21% for maximal peak density (not measured with

our software’s version) and 48% for AUC. On the other hand, Hinrichs et al.(187) had

correlated the PT pre- and post-intervention with the ABI (r = −0.53, p = 0.0081) whilst the

peak density (PD) and the AUC failed to demonstrate significant correlation with improved

ABI.(187) Consistently, our results have effectively shown an increment of the AUC of

2832±5053 (p<0.001), but without any prediction power on clinical outcomes. AUC/s, that

intended to correct the effect of the possible variability in duration of the angiographic run,

did not demonstrate predictive value on ulcer healing. On the contrary, the PT>5.2 s

(Sensitivity: 43.7%, Specificity: 76.9%, p:0.002) and the ∆PT>1.5s (Sensitivity: 40.9%,

Specificity: 80.8%, p:0.001) showed clinical relevance in our study, as Hinrichs had suggested.

However, a ∆MTT>1.7 s (OR: 3.21; 95% CI: 1.23-8,42, p:0.017) and an AT>6 s (OR: 2.64;

95% CI: 1.08-6.42, p:0.033) have appeared in the multivariate analysis as better predicting

factors for a TTH<30 days than PT>5.2 s (OR: 2.42; 95% CI: 0.98-6.02, p:0.056) or ∆PT>1.5

s (OR: 3.067; 95% CI: 1.19-7.88, p:0.02).

Another parameter measured by Reekers et al.(183) was the capillary resistance index.

We could not measure this parameter in our study because it is calculated from the maximal

peak density before and after the administration of tolazoline (a non-selective competitive µ-

adrenergic receptor antagonist).

We agree that intensity related parameters (considered as those parameters measuring

the blood volume passing through the foot) are probably measuring both micro- and

macrocirculation;(183) but the shape related parameters (PT, MTT, W, WS) are possibly more

dependent on the tissue hemodynamics and measuring then the microcirculation.(181) Hereby,

124

two conclusions can be extracted: they are of no less importance(183) and they are probably

more precise since they depend solely on the 90 % of the pixels in the ROI.(183) The latter

fact is indeed also applicable to the intensity related parameters, so it is plausible that AT>6s

is one of the independent predicting factors for the healing of ulcers. Along the same lines,

there is the finding that the ATS (analyzing merely the macro-circulation) appears to be

correlated with AT, AUC/s and AUC (sr:0.41, sr:0.45 and sr:0.31, respectively), with a mild-

to-moderate degree of correlation maybe due to the fact that they are still based on the 90%

measuring of the microcirculation.

Although PA contains microvascular information, it could asses the macrovascular

information. The application of the PA technology to the main arteries in order to explore the

effect on the big vessels had been explored by Kostrzewa et al.(188) They used a similar

technique to the PA (the parameter color coding), which allows to further interpret DSA series

and better visualize contrast medium dynamics through a particular stenosis or occlusion.

Nevertheless, they could not show a strong correlation between parametric color coding and

the clinical parameters of ABI and the ultrasound peak systolic velocity ratio.

6.2. Internal and external validity

The number of deaths and of cases with missing data before healing was not large

enough to be a potential source of confounding factors. Even if we considered all missing

cases as bad outcomes in healing (as we did in an out of study analysis) the results would be

similar in terms of significance and in main outcomes. Regarding the study population, we

can disclose that the kind of patients included and treated were the typical CLI patients

found in a diabetic foot clinic, patients with high prevalence of DM and multiple comorbidi-

ties. Since the center did not have hemodialysis facilities, the prevalence of ESRD in our

study population (9.2%) was similar to the general prevalence of ESRD, with the preva-

lence of CKD stages 3 to 5 in Europe being of 11.86%.(190)

The WIfI classification in the 30-days follow-up wass in the 96% in stages 1 or 2.

Subsequently we expected the general behaviour of the clinical predictions on those stages.

Thus, the likelihood of wound healing at 1 year was found to be 94.1±2.0% for stage 1

wounds,(21) which is similar to the 96.6% limb salvage found in our study. The same

consistance was observed in the validation of Cull et al.,(22) where the WIfI clinical stages

1 and 2 showed to be predictive of 1-year limb amputations of 3% and 10%, respectively.

125

Another potential factor affecting the results was the particularly aggressive treatment stra-

tegy for revascularization undertaken in our study center. The relatively low percentage of

stent and drug eluting technologies is a consequence of the bailout strategy for both. They

are usually applied on a clearly bad result on a first treatment or in an early failure of EVT

with clinical relevance. Comparing our results with the OPG described by Conte et al.(165),

our series were significantly better. While the OPG are 71%, 84% and 80% for Amputation

Free Survival (AFS), (freedom from amputation and survival at 1 year), our results at 6

months were 92.4%, 96.6% and 95.8%, respectively. The latter comparison is made with the

general population OPG, but our study population may better reflect an anatomical high risk

(78.2% of patients presenting infrapopliteal lesions in TASC) in which the OPG for the AFS

is 68%. Other results on clinical outcomes were reported by Egorova et al.(139), with AFS

at 1 year (2007) of 48-81% and the rates of major amputations ranging from 10 to 38%

(freedom from amputation of 62-90%), both much higher than the rates in our study.

A limitation that we should considered is that the iodinated contrast agent is a liquid

that advances by dilution in the bloodstream and not in phases (like the behavior of a gas, for

example). This implies that there will be an overlap between the speed of the blood flow and

the speed of dilution of the contrast agent, resulting in a slight shorter arrival speed and a

lower wash-in rate than the real blood and oxygen exchange rate. It has been observed in the

same way that iodinated contrast agents are not hemodynamically inert when explored on

coronary blood flow, they induce a flow reduction followed by a hyperemic response.(191) A

last consideration on the properties of the contrast medium on the blood stream, is that during

the first pass of the contrast, there is a diffusion into the interstitial space, which may suppose

a non-controlled disturbance of the measured contrast density.(191)

6.3. Interpretation of the results, possible confounding factors

and clinical translation. The results may seem paradoxical since larger perfusion times appear to be related to

better clinical outcomes. However, all of the PA parameters are the mean value of the curve

on each pixel on the ROI and the vast majority of pixels of the ROI are in the tissue rather

recording the flow in the big vessels. Therefore, we should shift our mind from the

hemodynamics in the large vessels to flow in a later stage in the tissue. In the tissue there are

the small vessels – not visible to the human naked eye – that define the shape of the PA curve.

As it can be seen in figure 34, the vast majority of vessels in a selected ROI will rely on

microcirculation rather than on big arteries that could be identified by conventional

126

angiography. Subsequently PA, which works with a bidimensional image and tridimensional

information, is focused on the study of small vessels with interferences from the big vessels.

It is noteworthy that all PA parameters (except the AUC) measure the shape of the curve in

terms of time and independently from the absolute contrast density values (at least in the PA

software version used in the current study). Thus, it would be intuitively consistent to consider

the flow behavior in the foot rather than the total amount of contrast (blood) arriving to the

foot.

Although an AT > 6 seconds appeared to predict a higher rate of healing before 30

days, there could exist a certain degree of vasospasm and endothelial dysfunction following

the aggression of the EVT. Especially in those technically complex cases that are more prone

to bad clinical outcomes, it could be justified (assuming the global improvement after the

EVT) to lower the mean AT among the whole study population. Notwithstanding, rather than

not acting, this hypothetic effect could event blur a real meaning of an AT>6 seconds. If the

degree of vasospasm or the endothelial dysfunction could be analyzed, it could improve the

diagnostic power of AT in predicting a good revascularization outcome. Vasospasm and acute

endothelial dysfunction should be considered as confounding factors since they would modify

the hemodynamics in the foot without occlusive lesions. This could be seen in the worsening

Figure 34. Plastination of microcirculation in the foot. The conventional DSA angio-

graphies (left) of CLI patients after treatment provide good assessments, but are far from

showing the complete microcirculatory tree existing after the plantar arch that can be seen

on the plastination model (right). CLI: Critical Limb Ischemia; DSA: Digital Subtraction

Angiography.

127

of the PA in some patients with very mild lesions at the beginning of the treatment, as it is

shown in figure PA5 (see annexes). Consequently, we could expect some grade of worsening

of the readouts just because of vasospasm due to the treatment itself. This fact could be a

reason why the ∆ parameters are not the most exact measures, and consequently the post-

treatment parameter shall become the more useful one.

6.4. Comparison with other diagnostic methods As suggested by Venermo et al.,(93) all diagnostic methods have their own strengths

and pitfalls. The clinician should then combine several methods to get the better information

about the patient prognostic and potential improvements of revascularization. An ideal

diagnostic method to serve as an intraoperative decision making should (1) be able to predict

clinical outcomes; (2) have no limitations due to calcifications; (3) non-invasive or not

requiring extra-intervention; and, (4) able to be performed during revascularization to assist

the peri-procedure decision making. Additionally, it should: (5) be reproducible; (6) easy to

do during the practice; (7) provide anatomical data; and, (8) provide histological tissue

information. Extending on the two latter properties, an ideal test should also be able to

measure separately the macrovascular and the microvascular disease and identify the location

of the culprit of the disease; or at least, even if a test would be sensible to both vascular beds

simultaneously, it should allow the identification the site of the effect measured. The

histological information is referred to the distribution of the perfusion to the ulcers or towards

the specific regions of tissue that we intend to treat. In table 22 we summarized the existing

diagnostic methods with an approach to describe the main attributes above mentioned. Among

all diagnostic methods, the PA appears to be a powerful tool, despite the availability of the

technology is not still widespread. Lack of access is a common situation for all novel

technologies while they are standardized and clinically tested.

The pressure-related quantitative diagnostic methods (ABI, AP and TP) present the

inconvenience of a high variability and lack of measurement precision becoming not very

enlightening for follow-up analysis, but they perform well for screening and for an initial

diagnosis. In addition, they could not assist during procedures due to the prolonged time

needed for a measure and the lack of histological or precise angiosome related improvement

of EVT. Despite TP pressure overcomes the limitation of the arterial wall stiffness due to

calcifications and its clinical significance is well-known, it still does not give objective

information about the perfusion to a concrete ulcer. On the other hand, the pressure-related

qualitative methods as PVRs and PPG do not differ from the conventional angiography in the

128

sense of being roughly estimated subjective values, and are not suitable for intraprocedural

assessment. Thus, they do not afford a threshold value, nor comparisons between patients or

interventions (inter-observer and intra-observer variations).

TcPO2 has two main drawbacks, the need of a temperature-controlled environment

and the time-consuming nature of the test. However, it is a well-known and clinically proven

test with a clear threshold of 25 mmHg with high sensitivity (85%) and specificity (92%) for

ulcers to heal. The PA parameters did not result in better predicting power, since the AT>6

seconds and the ∆MTT>1.7 seconds showed a sensitivity of 64.8% and 42.4%, respectively;

while their specificities were 59% and 81.7%. However, the TcPO2 cannot be used during

procedures nor distinguish among different angiosomes or specific histological measures.

This latter issue has been solved by the tissue oxygen saturation mapping, which is able to

measure quickly (only 10 seconds/point) the oxygen saturation through all the tissue. It is then

a really promising method with the theoretical ability of being sensible to detect vasospasm

and endothelial dysfunction, but it is still clinically unproven. Last but not least, the mild

correlation between TcPO2 and AT (sr=0.23, p=0.029) could indicate that both methods are

not measuring exactly the same pathophysiological process. Otherwise, if it were the case that

they both measure perfusion, the two methods will have to be applied at different times

(intraprocedural for AT and ambulatory for TcPO2). In the latter hypothesis, the lower

sensitivity and specificity of the PA could be explained by a subgroup of patients that having

had the same angiographic result, an eventual greater post-procedure endothelial dysfunction

did not lead to a good clinical result as the angiographic result could predict.

Based on a similar principle of the TcPO2 mapping, there are several methods able to

assess the skin perfusion: Hyperspectral imaging (HI), O2C, MOXYs and ICG-FI. None of

them has any large clinical follow-up study testing patient-centered clinical outcomes. Only

HI has showed predicting potential for ulcer healing (sensitivity=80% and PPV=90%), as

Nouvong et al.(106) reported. Comparatively, those are better results than those obtained by

TcPO2 and PA, and further HI could be applied during the procedure. However, HI is sensibly

limited by common factors such as deep tissue ischemia or infection (osteomyelitis), wounds,

scars or callus, because it only measures surface (1 cm) perfusion. Nonetheless, it could have

its own meaningful use in the ischemic assessment and procedure decision making;

particularly in prevention and follow-up.

Another similar measuring method is the O2C, able to measure not only the oxygen

saturation until 8 mm, but also hemoglobin and the blood flow, and it could be used during

procedures. However, the only clinical data available with from this technique aimed to

predict the prognosis of amputation levels.

129

MOXYs could be hard to standardize due to its invasiveness, incompatibility with

infection or ulcers, and the thin surface able to be measured (from 2 to 4 mm). There are no

studies measuring clinical outcomes although it is sensible to revascularization.

To end with surface perfusion measuring methods there is the ICG-FI. Despite the

exploration lasts 5 minutes and that it needs extra intervention procedures, it requires special

environment and temperature conditions. This method showed strong correlations with TP

and TcPO2, but clinical follow-up data is not yet available.

As of yet, none of the above-mentioned methods gives actual anatomical information.

The linkage between ulcer perfusion and the anatomical distribution of steno-occlusions is

partially solved by the angiosome theory. Unfortunately, no method could give any more

personalized assessment than attaching to the theoretical distribution of the angiosomes

without the presence of anatomical variations of the tibial arteries. The PA measurement,

however, is based on the ROI traced just adapted to the target tissue (ulcer). Furthermore, the

PA will always be associated with a conventional angiography which does have the anatomical

information of the steno-occlusions in the limb. The clue then is to be able to categorize the

subjective information about the lesion’s distribution contained in the angiography, such as it

is the aim of the ATS (despite that no strong correlation was found with TcPO2 or PA). In that

case we could be able to compare patients and set an anatomical revascularization threshold.

The other diagnostic methods containing anatomical information are DUS, MSOT

(multispectral optoacustic tomography), CTA and MRI angiography; with just the latter being

able to measure the tissue perfusion (but without any clinical outcomes data). All of them,

specially the three last ones, are confined to the treatment planning period because they cannot

be performed during the EVT. DUS on the other hand could be performed periprocedurally,

it is highly correlated with angiography and gives real hemodynamic information; however,

no correlation with the microcirculation state or the clinical long-term outcomes are known.

Furthermore, it is very time-consuming and operator dependent, which becomes a problem to

compare among patients. A new modality of ultrasound imaging is the MSOT, and while it

could give a trustworthy vasculature picture of 14 mm-depth, no clinical or feasibility studies

are yet available.

Considering the big picture among current diagnostic methods, the results obtained

and the clinical applicability of PA, we constructed an algorithm (figure 35) to integrate this

information and make it applicable into routine clinical practice. A particular consideration,

when the described threshold values of PA are not reached, is that it relays on the clinical

judgement of the attending physician. In the case that the intended procedure for

revascularization is achieved, the viability of the limb should be first considered, or the patient

130

Able

to p

redi

ct

clin

ical

out

com

es

Lim

itatio

ns d

ue to

calc

ifica

tions

Inva

sive

Peri

-pro

cedu

ral

min

/mea

sure

Anat

omic

al

info

rmat

ion

His

tolo

gica

l

info

rmat

ion

ABI + ++ No No 10 No No

AP + ++ No No 10 No No

PVRs - - No No 10 No No

PPG - - No No 10 No No

TP ++ - No No 10 No No

TcPO2 ++ - No No 20-40 No No

TcPO2 mapping N/A - No N/A <5 No Yes

Hyperspectral imaging ++ - No Yes <5 No Yes

O2C N/A - No No N/A No Yes

MOXYs N/A - Yes Yes N/A No No

ICG-FI N/A - No No 15 No Yes

PA ++ - No Yes <5 No Yes

DUS N/A + No Yes 30-40 Yes Yes

MSOT N/A N/A No No N/A Yes Yes

CT angiography N/A ++ Yes No 20 Yes No

MRI angiography N/A ++ No No 30-60 Yes Yes

Table 22. Compared diagnostic methods. AP: ankle pressure; DUS: duplex ultrasound; ICG-FI: indocyanine-green fluorescence

imaging; MOXYs: micro-oxygen sensors; MSOT: multispectral optoacustic tomography; O2C: oxygen-2C; PA: perfusion angiography;

PPG: photopletysmography; PVRs: pulse volume recordings; TP: toe pressure.

131

should be rescheduled for a reintervention, even applying vasodilators as a bailout therapy.

On the other hand, if the intended procedure for the EVT could not be achieved, it would

depend on technical issues of the procedure to decide rescheduling the patient or reconsidering

the risk of a reintervention over the one of an above the ankle amputation.

6.5. Secondary results

6.5.1. Disease burden anatomical classification.

The current most used classification to describe the steno-occlusive lesions in PAD

is the TASC, but it has several pitfalls. To begin with, it is not easy to remember, its aim is

focused on the election of the revascularization technique (open surgery or endovascular)

rather than the hemodynamic description of a limb; furthermore, the classification does not

raise a single value for the whole limb. Concretely in the BTK TASC classification, the

concept of the target vessel is ambiguous. Beside the lesions of the other arteries, the target

vessel in the current study had been chosen as the angiosome related vessel, since this is the

vessel with the major interest to treat. Nevertheless, anatomical variations should be

considered and in patients with rest pain the target tibial will be the one with less severe

lesions. In the post-treatment angiography, the target vessel had been chosen as the one with

Figure 35. Decision algorithm to apply PA information during the intervention. AT: Arrival Time; CLI:

Critical Limb Ischemia; EVT: Endovascular Treatment; MTT: Mean Transit Time; PA: Perfusion angio-

graphy.

132

the best angiographic result. Therefore, it is hard to clearly determine a case to a BTK

TASC C or D, in the sense of the amount of calcium needed to consider a D classification.

With the herein described context, without any better performance of other systems like Bo-

llinger’s(192) or Graziani’s(193,194) scores, a simpler anatomic classification could be

really helpful to describe, compare and even set a revascularization goal. Better correlation

with the clinical outcome was demonstrated by Bargellini et al.(194), who compared JVSC

foot and calf scores with TASC and Graziani’s classification and found that only the JVSC

foot scores (but not the calf score) where related to better healing time; both with lower pre-

procedural (mean score, 7 in healed patients and 7.8 in non-healed patients; p = .049) and

postprocedural (mean score, 5.5 in healed patients and 6.3 in nonhealed patients; p = .047).

Given that the territories that describe the ATS and the JVSC foot score are in fact comple-

mentary and both systems are numerical, this could suppose an advantage for its clinical ap-

plication.

Ideally the classification to be used should be easy to remember and apply, based on

hemodynamic criteria, avoiding as much as possible arbitrariness in the definition and with

a certain degree of clinical relevance. This has been the aim of the herein described Abano

Terme Score (ATS). ATS appeared to be an easy to describe and to remember score system

that is moderately correlated with the AT and the AUC/s when applying the ATS for the

whole limb. Those were a Spearman rho correlation of 0.41 for the AT and -0.45 for the

AUC/s. Although they are not excellent, they are better than the TASC classification. Furt-

hermore, although it has not been related to the primary endpoint of this study (TTH<30

days), the ATS is actually sensible to the EVT. The ATS mean score pre-EVT was 10.6±5.8

and post-EVT was 1.43±2.3 (p<0.001). A good point of the ATS is that it would be simple

and always applicable to every patient undergoing a conventional angiography, without re-

quiring any extra intervention or technology use.

Recently though, an expert global panel as part of the Global Vascular Guidelines on

Chronic Limb Threatening Ischaemia designed the GLASS score. GLASS is part of the

Patient, Limb, Anatomy paradigm presented in the Global Vascular Guidelines. GLASS

involves choosing the target artery pathway for endovascular revascularization from the

origin of the SFA (common and deep femoral disease are considered part of inflow and

considered corrected before more distal revascularization) to the foot with the aim of

establishing inline flow. Disease severity in FP and infrapopliteal (IP) segments are graded

separately on features including duration of disease, stenosis or occlusion, and level of

calcification. FP and IP grades are combined within a matrix to determine GLASS stage.

The GLASS stage is believed to likely correlate with endo-vascular immediate technical

133

success rates and 12-month limb-based patency. GLASS has been presented at several

congresses and symposia (European Society of Vascular Surgery, Lyon, France in 2017, the

Society of Vascular Surgery VAM in San Diego in 2017, and in Boston in 2018, and in the

Charing Cross Symposium in London in 2018). In a recent study Kodama et al.(18) demon-

strated the GLASS can predict technical failure, AFS and major adverse limb events (MALE). How-

ever, only the baseline angiography was used, so no treatment endpoint could be extracted from

those studies. On the other hand, the GLASS feasibility study,(32) showed inter-observer variation

in applying the classification system. Further trials and investigations should be done to select

the most powerful score and which one is more relevant to clinical daily practice.

6.5.2. TcPO2 correlation with perfusion angiography

The decrease of the TCPO2 at the 6-months follow-up from 50.9±11.9 mmHg to

32.2±17.2 mmHg (just slightly higher mean than the one before the revascularization) seems

to indicate an important restenosis rate. Nonetheless, this has no clinical implications nor

could be compared to the Objective Performance Goals (OPG) defined by Conte et al.(159)

because it is not directly a stenosis nor an occlusion. The TCPO2 showed a mild but statisti-

cal correlation with the AT (sr=0.23, p=0.029), which seems consistently indicate that both

tests measure the oxygen arrival to the tissues. Additionally, the TCPO2 was correlated with

the ATS for BTK (sr=0.25, p=0.020) and for the whole limb (sr=0.24, p=0.023).

6.5.3. Effects of infra-popliteal anatomical variations

There were no significant associations between the presence of an anatomical

variation in tibial arteries and rest pain clinical presentation. However, since a tendency for

this association had been seen (17.7% among the patients with ulcers and 29.7% in those with

rest pain, p= 0.146), it worth giving an extra attention to the possibility of an anatomical

variation when treating a patient with rest pain. A possible explanation for this is the fact that

an anatomical variation usually implies the existence of an hypoplasic artery. Subsequently,

the remaining two arteries are responsible for a higher proportion of whole foot perfusions;

being one over two instead of one over three arteries. In this situation, a steno-occlusion in

one infra-popliteal artery will affect one half rather than one third of the blood supply, which

For more information see section 11.5 about the submitted article to the European Journal of Vascular & Endovascular Surgery: Novel Abano Terme Scores and its application in the treatment of critical limb threatening ischemia.

134

supposes a more severe ischemia present in the rest pain. Finally, there is a higher risk of

failure for revascularization when an anatomical variation is not well noticed before

revascularization. It is striking that we detected significantly more anatomical variations than

those described in the literature, where the proportion is from 9 to 12 %. It is not easy though

to precisely describe anatomical variations over occluded and extremely diseased arteries.

6.5.4. Effects of a cropped plantar arch

A special consideration of the application of the PA should be made for patients with

a previous cropped plantar arch. In those patients PT, WS, W and MTT appeared to be

statistically different. This endorses the hemodynamic main importance of the plantar arch

patency for the tissue perfusion in the foot.

6.6. Limitations and further possibilities We have had a lack of standardization of the projections and the PA protocol in the

study, since the objective was to explore the technique for daily practice. The same problem

has occurred with the proportion of missing cases during follow-up. Over those data we made

exploratory statistics to test the PA predicting value of clinical results after EVT. This

approach could sometimes produce random findings.(195)

Due to the retrospective nature of the data collection, the vast majority of patients have

not clearly recorded the active or former smoking habits. Thus, the lack of that potentially

significant information could be hiding another confusion factor.

Another limitation is that the follow-up only lasted 6 months, while the suggested

duration for follow-up of CLI revascularizations is 1 year;(1) however in our study, the effects

of revascularization were studied in a shorter period of time (30 days).

A particular characteristic for our studied population is the divergence in the patient

centered outcomes, which are significantly better than those described in other large series

and RCTs. It would be probably useful to compare the results in this study with other less

aggressively treated patients, and compare the significance of the PA among patients in all

stages of disease burden and degree of treatment.

From a conceptual point of view, we should understand the differences between the

blood flow through the big arteries, as those in which we could do a balloon dilatation (dorsalis

pedis, lateral plantar, plantar arch, lateral tarsal, etc), and in the small arteries, whose blood

flow is affected by MAC. It is at tissue level where the oxygen and the nutrients carried in the

blood are ultimately needed. With this picture in mind we can easily imagine that every

135

diagnostic method will be better exploring one or another step-in perfusion, as the complete

process of oxygen and nutrients delivered to the tissues. The real implication in CLI clinical

outcomes of Mönckeberg’s disease and the management of SHPT with phosphate binders,

active vitamin D analogs and Ca2+ mimetics should be explored. In other words, the

management through other strategies rather than the interventional one should be investigated,

such as the use of novel specific drugs or cellular therapies to improve the microvascular

circulation and ulcer tissue angiogenesis.

It would be really useful to compare the ATS with other CLI endpoints (like limb

salvage or freedom from TLR) and other well-known diagnostic methods like TcPO2 and the

duplex ultrasound. Probably a global limb threshold could predict limb salvage and the ATS

for BTK could predict outcomes after a femoropopliteal bypass (as a bypass outflow

quantification). The ATS is simple to describe and it is easy to obtain a numeric value to the

PAD disease burden in the limb; both considering the limb entirely or above- or below-the-

knee separately.

The PA software itself could also be substantially improved by making a software not

distortable by movement, or being able to make an automated selection over the course of a

single vessel for specific measures. It could even be equipped with machine learning

processing to develop specific treatment recommendations and with predictions of tissue

perfusion improvement.

137

7. CONCLUSIONS

139

The conclusions of this thesis are:

1. The PA parameters are sensible to revascularization of CLI patients and to the

real perfusion of the foot.

2. The threshold values of PA parameters to predict clinical improvement are:

2a. The best predicting factors for the healing of an ulcer in less than 30

days were AT>6 seconds in the post-treatment PA run and presenting an

∆MTT > 1.7 seconds (or an MTT > 4.1 seconds in the post-treatment run).

2b. No significative translation from perfusion angiography parameters to

clinical outcomes (freedom from reintervention) were seen in Rutherford

3-4 patients.

3. A classification hereby called “Abano Terme Score” appeared to be simple

and useful for clinical practice, being mild-moderately correlated not only with

AT, AUC and AUC/s, but also with the TcPO2, which also showed the same

strength correlation with AT.

4. The arrival time, as the parameter of the perfusion angiography, showed a mild

but statistical correlation with the TCPO2.

5. The parameters related with the perfusion curve shape (PT, WS, W and MTT)

were significantly related with the presence of an incomplete plantar arch.

141

8. REFERENCES

143

1. Mills JL, Conte MS, Armstrong DG, Pomposelli FB, Schanzer A, Sidawy AN, et al. The society for vascular surgery lower extremity threatened limb classification system: Risk stratification based on Wound, Ischemia, and foot Infection (WIfI). J Vasc Surg. 2014;

2. Hiatt WR, Goldstone J, Smith SC, McDermott M, Moneta G, Oka R, et al. Atherosclerotic peripheral vascular disease symposium II: Nomenclature for vascular diseases. Circulation. 2008;118(25):2826–9.

3. Criqui MH, Aboyans V. Epidemiology of Peripheral Artery Disease. Circ Res. 2015; 4. Bell PRF, Charlesworth D, DePalma RG, Eastcott HHG, Eklöf B, Jamieson CW et

al. The definition of critical ischemia of a limb. Working Party of the International Vascular Symposium. Br J Surg. 1982;69.

5. Fowkes FGR, Rudan D, Rudan I, Aboyans V, Denenberg JO, McDermott MM, et al. Comparison of global estimates of prevalence and risk factors for peripheral artery disease in 2000 and 2010: A systematic review and analysis. Lancet [Internet]. 2013;382(9901):1329–40. Available from: http://dx.doi.org/10.1016/S0140-6736(13)61249-0

6. Alahdab F, Wang AT, Elraiyah TA, Malgor RD, Rizvi AZ, Lane MA, et al. A systematic review for the screening for peripheral arterial disease in asymptomatic patients. J Vasc Surg [Internet]. 2015;61(3):42S-53S. Available from: http://dx.doi.org/10.1016/j.jvs.2014.12.008

7. McGee DL. Update on Some Epidemiologic Features of Intermittent Claudication: The Framingham Study. J Am Geriatr Soc. 1985;33(1):13–8.

8. Beckman JA, Creager MA. Critical limb ischemia and intermediate-term survival. JACC Cardiovasc Interv [Internet]. 2014;7(12):1450–2. Available from: http://dx.doi.org/10.1016/j.jcin.2014.07.012

9. Hooi JD, Kester AD, Stoffers HE, Overdijk MM, van Ree JW, Knottnerus JA. Incidence of and risk factors for asymptomatic peripheral arterial occlusive disease: a longitudinal study. Am J Epidemiol. 2001 Apr;153(7):666–72.

10. Murabito JM, Evans JC, Nieto K, Larson MG, Levy D, Wilson PW f. Prevalence and clinical correlates of peripheral arterial disease in the Framingham Offspring Study. Am Heart J. 2002 Jun;143(6):961–5.

11. Meijer WT, Hoes AW, Rutgers D, Bots ML, Hofman A, Grobbee DE. Peripheral arterial disease in the elderly: The Rotterdam Study. Arterioscler Thromb Vasc Biol. 1998 Feb;18(2):185–92.

12. Gallotta G, Iazzetta N, Milan G, Ruocco A, Napoli C, Postiglione A. Prevalence of peripheral arterial disease in an elderly rural population of southern Italy. Gerontology. 1997;43(5):289–95.

13. Newman AB, Siscovick DS, Manolio TA, Polak J, Fried LP, Borhani NO, et al. Ankle-arm index as a marker of atherosclerosis in the Cardiovascular Health Study. Cardiovascular Heart Study (CHS) Collaborative Research Group. Circulation. 1993 Sep;88(3):837–45.

14. Second European Consensus Document on chronic critical leg ischemia. Eur J Vasc Surg. 1992 May;6 Suppl A:1–32.

15. Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes FGR, et al. Inter-Society Consensus for the management of peripheral arterial disease (TASC II). J Vasc Surg. 2007;26(2):82–157.

16. Vallabhaneni R, Kalbaugh CA, Kouri A, Farber MA, Marston WA. Current accepted hemodynamic criteria for critical limb ischemia do not accurately stratify patients at high risk for limb loss. J Vasc Surg [Internet]. 2016;63(1):105–13. Available from: http://dx.doi.org/10.1016/j.jvs.2015.07.095

17. Armstrong EJ, Alam S, Henao S, Lee AC, DeRubertis BG, Montero-Baker M, et al. Multidisciplinary Care for Critical Limb Ischemia: Current Gaps and Opportunities

144

for Improvement. J Endovasc Ther [Internet]. 2019;152660281982659. Available from: http://journals.sagepub.com/doi/10.1177/1526602819826593

18. Robinson WP, Loretz L, Hanesian C, Flahive J, Bostrom J, Lunig N, et al. Society for Vascular Surgery Wound, Ischemia, foot Infection (WIfI) score correlates with the intensity of multimodal limb treatment and patient-centered outcomes in patients with threatened limbs managed in a limb preservation center. J Vasc Surg [Internet]. 2017;66(2):488-498.e2. Available from: http://dx.doi.org/10.1016/j.jvs.2017.01.063

19. Darling JD, McCallum JC, Soden PA, Meng Y, Wyers MC, Hamdan AD, et al. Predictive ability of the Society for Vascular Surgery Wound, Ischemia, and foot Infection (WIfI) classification system following infrapopliteal endovascular interventions for critical limb ischemia. In: Journal of Vascular Surgery. 2016. p. 616–22.

20. Beropoulis E, Stavroulakis K, Schwindt A, Stachmann A, Torsello G, Bisdas T. Validation of the Wound, Ischemia, foot Infection (WIfI) classification system in nondiabetic patients treated by endovascular means for critical limb ischemia. J Vasc Surg [Internet]. 2016;64(1):95–103. Available from: http://dx.doi.org/10.1016/j.jvs.2016.01.040

21. Hicks CW, Canner JK, Mathioudakis N, Sherman R, Malas MB, Black JH, et al. The Society for Vascular Surgery Wound, Ischemia, and foot Infection (WIfI) classification independently predicts wound healing in diabetic foot ulcers. J Vasc Surg [Internet]. 2018;68(4):1096–103. Available from: https://doi.org/10.1016/j.jvs.2017.12.079

22. Cull DL, Manos G, Hartley MC, Taylor SM, Langan EM, Eidt JF, et al. An early validation of the Society for Vascular Surgery Lower Extremity Threatened Limb Classification System. J Vasc Surg [Internet]. 2014;60(6):1535–42. Available from: http://dx.doi.org/10.1016/j.jvs.2014.08.107

23. Zhan LX, Branco BC, Armstrong DG, Mills JL. The Society for Vascular Surgery lower extremity threatened limb classification system based on Wound, Ischemia, and foot Infection (WIfI) correlates with risk of major amputation and time to wound healing. J Vasc Surg [Internet]. 2015;61(4):939–44. Available from: http://dx.doi.org/10.1016/j.jvs.2014.11.045

24. Hicks CW, Canner JK, Karagozlu H, Mathioudakis N, Sherman RL, Black JH, et al. The Society for Vascular Surgery Wound, Ischemia, and foot Infection (WIfI) classification system correlates with cost of care for diabetic foot ulcers treated in a multidisciplinary setting. J Vasc Surg [Internet]. 2018;67(5):1455–62. Available from: https://doi.org/10.1016/j.jvs.2017.08.090

25. Mathioudakis N, Hicks CW, Canner JK, Sherman R, Malas MB, Black JH, et al. The Society for Vascular Surgery Wound, Ischemia, and foot Infection (WIfI) classification system predicts wound healing but not major amputation in patients with diabetic foot ulcers treated in a multidisciplinary setting. J Vasc Surg [Internet]. 2018;68(4):1096–103. Available from: http://dx.doi.org/10.1016/j.jvs.2016.12.123

26. WHO. Global status report on noncommunicable diseases 2014. World Health. 2014;176.

27. Selvin E, Marinopoulos S, Berkenblit G, Rami T, Brancati FL, Powe NR, et al. Meta-analysis: glycosylated hemoglobin and cardiovascular disease in diabetes mellitus. Ann Intern Med. 2004 Sep;141(6):421–31.

28. Jude EB, Oyibo SO, Chalmers N, Boulton AJ. Peripheral arterial disease in diabetic and nondiabetic patients: a comparison of severity and outcome. Diabetes Care. 2001 Aug;24(8):1433–7.

29. Jonason T, Bergstrom R. Cessation of smoking in patients with intermittent claudication. Effects on the risk of peripheral vascular complications, myocardial infarction and mortality. Acta Med Scand. 1987;221(3):253–60.

145

30. He Y, Lam TH, Jiang B, Wang J, Sai X, Fan L, et al. Passive smoking and risk of peripheral arterial disease and ischemic stroke in Chinese women who never smoked. Circulation. 2008 Oct;118(15):1535–40.

31. Fowkes FG, Housley E, Riemersma RA, Macintyre CC, Cawood EH, Prescott RJ, et al. Smoking, lipids, glucose intolerance, and blood pressure as risk factors for peripheral atherosclerosis compared with ischemic heart disease in the Edinburgh Artery Study. Am J Epidemiol. 1992 Feb;135(4):331–40.

32. Joosten MM, Pai JK, Bertoia ML, Gansevoort RT, Bakker SJL, Cooke JP, et al. beta2-microglobulin, cystatin C, and creatinine and risk of symptomatic peripheral artery disease. J Am Heart Assoc. 2014 Jun;3(4).

33. Romero JM, Bover J, Fite J, Bellmunt S, Dilmé JF, Camacho M, et al. The modification of diet in renal disease 4-calculated glomerular filtration rate is a better prognostic factor of cardiovascular events than classical cardiovascular risk factors in patients with peripheral arterial disease. J Vasc Surg [Internet]. 2012;56(5):1324–30. Available from: http://dx.doi.org/10.1016/j.jvs.2012.04.049

34. Selvin E, Erlinger TP. Prevalence of and risk factors for peripheral arterial disease in the United States: results from the National Health and Nutrition Examination Survey, 1999-2000. Circulation. 2004 Aug;110(6):738–43.

35. Murabito JM, D’Agostino RB, Silbershatz H, Wilson WF. Intermittent claudication. A risk profile from The Framingham Heart Study. Circulation. 1997 Jul;96(1):44–9.

36. Natarajan S, Glick H, Criqui M, Horowitz D, Lipsitz SR, Kinosian B. Cholesterol measures to identify and treat individuals at risk for coronary heart disease. Am J Prev Med. 2003 Jul;25(1):50–7.

37. Kannel WB, McGee DL. Update on some epidemiologic features of intermittent claudication: the Framingham Study. J Am Geriatr Soc. 1985 Jan;33(1):13–8.

38. Katsilambros NL, Tsapogas PC, Arvanitis MP, Tritos NA, Alexiou ZP, Rigas KL. Risk factors for lower extremity arterial disease in non-insulin-dependent diabetic persons. Diabet Med. 1996 Mar;13(3):243–6.

39. Vogt MT, Cauley JA, Kuller LH, Hulley SB. Prevalence and correlates of lower extremity arterial disease in elderly women. Am J Epidemiol. 1993 Mar;137(5):559–68.

40. Fabsitz RR, Sidawy AN, Go O, Lee ET, Welty TK, Devereux RB, et al. Prevalence of peripheral arterial disease and associated risk factors in American Indians: the Strong Heart Study. Am J Epidemiol. 1999 Feb;149(4):330–8.

41. Curb JD, Masaki K, Rodriguez BL, Abbott RD, Burchfiel CM, Chen R, et al. Peripheral artery disease and cardiovascular risk factors in the elderly. The Honolulu Heart Program. Arterioscler Thromb Vasc Biol. 1996 Dec;16(12):1495–500.

42. Amrock SM, Weitzman M. Multiple biomarkers for mortality prediction in peripheral arterial disease. Vasc Med (United Kingdom). 2016;21(2):105–12.

43. Demirdal T, Sen P. The significance of neutrophil-lymphocyte ratio, platelet-lymphocyte ratio and lymphocyte-monocyte ratio in predicting peripheral arterial disease, peripheral neuropathy, osteomyelitis and amputation in diabetic foot infection. Diabetes Res Clin Pract [Internet]. 2018;144:118–25. Available from: https://doi.org/10.1016/j.diabres.2018.08.009

44. Suades Soler R. Role of microparticles in atherothrombosis. Universitat Pompeu Fabra; 2014.

45. Cubedo J, Vilahur G, Casaní L, Mendieta G, Gómez-Jabalera E, Juan-Babot O, et al. Targeting the molecular mechanisms of ischemic damage: Protective effects of alpha-crystallin-B. Int J Cardiol. 2016;215.

46. Bhatt DL. International Prevalence, Recognition, and Treatment of Cardiovascular Risk Factors in Outpatients With Atherothrombosis. Jama [Internet]. 2006;295(2):180. Available from:

146

http://jama.jamanetwork.com/article.aspx?doi=10.1001/jama.295.2.180 47. Coccheri S. Distribution of symptomatic atherothrombosis and influence of

atherosclerotic disease on risk of secondary ischaemic events: Results from CAPRIE. Eur Hear J. 1998;19(Suppl):227.

48. Chapman MJ. From pathophysiology to targeted therapy for atherothrombosis: A role for the combination of statin and aspirin in secondary prevention. Pharmacol Ther. 2007;113(1):184–96.

49. Lozano R, Naghavi M, Foreman K, Lim S, Shibuya K, Aboyans V, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: A systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380(9859):2095–128.

50. Mathers CD, Loncar D. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med. 2006;3(11):2011–30.

51. Rocha-Singh KJ, Zeller T, Jaff MR. Peripheral arterial calcification: Prevalence, mechanism, detection, and clinical implications. Catheter Cardiovasc Interv. 2014;83(6):212–20.

52. Karwowski W, Naumnik B, Szczepanski M, Mysliwiec M. The mechanism of vascular calcification - a systematic review. Med Sci Monit. 2012 Jan;18(1):RA1-11.

53. Espitia O, Chatelais M, Steenman M, Charrier C, Maurel B, Georges S, et al. Implication of molecular vascular smooth muscle cell heterogeneity among arterial beds in arterial calcification. PLoS One. 2018;13(1):1–17.

54. Zheng G, Li Y, Huang H, Wang J, Hirayama A, Lin J. The effect of statin therapy on coronary plaque composition using virtual histology intravascular ultrasound: A meta-analysis. PLoS One. 2015;10(7):1–14.

55. Sage AP, Tintut Y, Demer LL. Regulatory Mechanisms in Atherosclerotic Calcification. Nat Rev Cardiol. 2010;7(9):528–36.

56. Shao J-S, Cheng S-L, Sadhu J, Towler DA. Inflammation and the osteogenic regulation of vascular calcification: a review and perspective. Hypertens (Dallas, Tex 1979). 2010 Mar;55(3):579–92.

57. Giachelli CM. Vascular calcification mechanisms. J Am Soc Nephrol. 2004 Dec;15(12):2959–64.

58. Tintut Y, Alfonso Z, Saini T, Radcliff K, Watson K, Boström K, et al. Multilineage Potential of Cells from the Artery Wall. Circulation. 2003;108(20):2505–10.

59. Nitta K. Vascular calcification in patients with chronic kidney disease. Ther Apher Dial. 2011 Dec;15(6):513–21.

60. Madhavan M V., Tarigopula M, Mintz GS, Maehara A, Stone GW, Généreux P. Coronary artery calcification: Pathogenesis and prognostic implications. J Am Coll Cardiol [Internet]. 2014;63(17):1703–14. Available from: http://dx.doi.org/10.1016/j.jacc.2014.01.017

61. Vliegenthart R, Oudkerk M, Hofman A, Oei HHS, Van Dijck W, Van Rooij FJA, et al. Coronary calcification improves cardiovascular risk prediction in the elderly. Circulation. 2005;112(4):572–7.

62. Abedin M, Tintut Y, Demer LL. Vascular calcification: Mechanisms and clinical ramifications. Arterioscler Thromb Vasc Biol. 2004;24(7):1161–70.

63. Davaine JM, Quillard T, Brion R, Lapérine O, Guyomarch B, Merlini T, et al. Osteoprotegerin, pericytes and bone-like vascular calcification are associated with carotid plaque stability. PLoS One. 2014;9(9):1–9.

64. Van Den Bouwhuijsen QJA, Bos D, Ikram MA, Hofman A, Krestin GP, Franco OH, et al. Coexistence of calcification, intraplaque hemorrhage and lipid core within the asymptomatic atherosclerotic carotid plaque: The Rotterdam study. Cerebrovasc Dis. 2015;39(5–6):319–24.

65. Herisson F, Heymann MF, Chétiveaux M, Charrier C, Battaglia S, Pilet P, et al.

147

Carotid and femoral atherosclerotic plaques show different morphology. Atherosclerosis. 2011;216(2):348–54.

66. Gouëffic Y, Davaine JM, Merlini T, Rimbert A, Hérisson F, Heymann MF, et al. Artériopathie et hétérogénéité artérielle. Rev Med Interne [Internet]. 2013;34(1):61–5. Available from: http://dx.doi.org/10.1016/j.revmed.2012.10.364

67. Davaine JM, Quillard T, Chatelais M, Guilbaud F, Brion R, Guyomarch B, et al. Bone like arterial calcification in femoral atherosclerotic lesions: Prevalence and role of osteoprotegerin and pericytes. Eur J Vasc Endovasc Surg. 2016;51(2):259–67.

68. Galan C, Jardín I, Dionisio N, Salido G, Rosado JA. Role of oxidant scavengers in the prevention of Ca2+homeostasis disorders. Molecules. 2010;15(10):7167–87.

69. Chertow GM, Burke SK, Raggi P, Chertow GM, Caputo G, Raggi P, et al. Sevelamer attenuates the progression of coronary and aortic calcification in hemodialysis patients. Kidney Int. 2002 Jul;62(1):245–52.

70. Kovesdy CP, Lu JL, Malakauskas SM, Andress DL, Kalantar-Zadeh K, Ahmadzadeh S. Paricalcitol versus ergocalciferol for secondary hyperparathyroidism in CKD stages 3 and 4: a randomized controlled trial. Am J Kidney Dis. 2012 Jan;59(1):58–66.

71. Pilz S, Iodice S, Zittermann A, Grant WB, Gandini S. Vitamin D status and mortality risk in CKD: a meta-analysis of prospective studies. Am J Kidney Dis. 2011 Sep;58(3):374–82.

72. Raggi P, Chertow GM, Torres PU, Csiky B, Naso A, Nossuli K, et al. The ADVANCE study: a randomized study to evaluate the effects of cinacalcet plus low-dose vitamin D on vascular calcification in patients on hemodialysis. Nephrol Dial Transplant. 2011 Apr;26(4):1327–39.

73. Cunningham J, Danese M, Olson K, Klassen P, Chertow GM. Effects of the calcimimetic cinacalcet HCl on cardiovascular disease, fracture, and health-related quality of life in secondary hyperparathyroidism. Kidney Int. 2005;68(4):1793–800.

74. Murray CJL, Lopez AD. Measuring the Global Burden of Disease. N Engl J Med [Internet]. 2013;369(5):448–57. Available from: http://www.nejm.org/doi/10.1056/NEJMra1201534

75. Johnson RC, Leopold JA, Loscalzo J. Vascular calcification: Pathobiological mechanisms and clinical implications. Circ Res. 2006;99(10):1044–59.

76. London GM, Guerin AP, Marchais SJ, Metivier F, Pannier B, Adda H. Arterial media calcification in end-stage renal disease: impact on all-cause and cardiovascular mortality. Nephrol Dial Transplant. 2003 Sep;18(9):1731–40.

77. Rennenberg RJMW, Kessels AGH, Schurgers LJ, Van Engelshoven JMA, De Leeuw PW, Kroon AA. Vascular calcifications as a marker of increased cardiovascular risk: A meta-analysis. Vasc Health Risk Manag. 2009;5:185–97.

78. Muraoka Y, Nakamura H, Ph D, Iwai T. Radiographic and Histological Studies of Calcification in the Media of Muscular-type Arteries. 2001;172:164–72.

79. Smith CD, Bilmen JG, Iqbal S, Robey S, Pereira M. Medial Artery Calcification as an Indicator of Diabetic Peripheral Vascular Disease. Foot Ankle Int [Internet]. 2008;29(2):185–90. Available from: http://journals.sagepub.com/doi/10.3113/FAI.2008.0185

80. Amos RS, Wright V. MÖNCKEBERG’S ARTERIOSCLEROSIS AND METABOLIC BONE DISEASE. Lancet [Internet]. 1980;August 2:248–9. Available from: papers2://publication/uuid/A62EB6D0-A2AA-4B62-BD20-D085815D257B

81. Everhart JE, Pettitt DJ, Knowler WC, Rose FA, Bennett PH. Medial arterial calcification and its association with mortality and complications of diabetes. Diabetologia. 1988 Jan;31(1):16–23.

82. Chantelau E, Lee KM, Jungblut R. Association of below-knee atherosclerosis to medial arterial calcification in diabetes mellitus. Diabetes Res Clin Pract.

148

1995;29(3):169–72. 83. Niskanen L, Siitonen O, Suhonen M, Uusitupa MI. Medial artery calcification

predicts cardiovascular mortality in patients with NIDDM. Diabetes Care. 1994 Nov;17(11):1252–6.

84. Füessl HS, Schälzky H, Schewe S, Frey KW, Goebel FD. [Pathogenesis and clinical significance of Mönckeberg medial calcinosis]. Klin Wochenschr [Internet]. 1985;63(5):211–6. Available from: http://www.ncbi.nlm.nih.gov/pubmed/3990163

85. FONTAINE R, KIM M, KIENY R. [Surgical treatment of peripheral circulation disorders]. Helv Chir Acta. 1954 Dec;21(5–6):499–533.

86. Rutherford RB, Flanigan DP, Gupta SK, Johnston KW, Karmody A, Whittemore AD, et al. Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg. 1986;4(1):80–94.

87. Rutherford RB, Baker JD, Ernst C, Johnston KW, Porter JM, Ahn S, et al. Recommended standards for reports dealing with lower extremity ischemia: Revised version. J Vasc Surg. 1997;26(3):517–38.

88. Lavery LA, Armstrong DG, Harkless LB. Classification of diabetic foot wounds. J Foot Ankle Surg [Internet]. 1996;35(6):528–31. Available from: http://dx.doi.org/10.1016/S1067-2516(96)80125-6

89. Jaff MR, White CJ, Hiatt WR, Fowkes GR, Dormandy J, Razavi M, et al. An Update on Methods for Revascularization and Expansion of the TASC Lesion Classification to Include Below-the-Knee Arteries: A Supplement to the Inter-Society Consensus for the Management of Peripheral Arterial Disease (TASC II). Vasc Med (United Kingdom). 2015;20(5):465–78.

90. Beard JD, Gaines PA, Loftus I. Vascular and Endovascular Surgery. Fifth edit. ELSEVIER, editor. 2014. 1–89 p.

91. Hirsch AT, Criqui MH, Treat-Jacobson D, Regensteiner JG, Creager MA, Olin JW, et al. Peripheral arterial disease detection, awareness, and treatment in primary care. JAMA. 2001 Sep;286(11):1317–24.

92. Bhasin N, Scott DJA. Ankle Brachial Pressure Index: Identifying cardiovascular risk and improving diagnostic accuracy. J R Soc Med. 2007;100(1):4–5.

93. Venermo M, Vikatmaa P, Terasaki H, Sugano N. Vascular laboratory for critical limb ischaemia. Scand J Surg. 2012;101(2):86–93.

94. Setacci C, Ricco JB. Guidelines for critical limb ischaemia and diabetic foot - Introduction. Eur J Vasc Endovasc Surg [Internet]. 2011;42(SUPPL. 2):S1–3. Available from: http://dx.doi.org/10.1016/S1078-5884(11)00715-5

95. Carter SA, Tate RB. Value of toe pulse waves in addition to systolic pressures in the assessment of the severity of peripheral arterial disease and critical limb ischemia. J Vasc Surg. 1996;24(2):258–65.

96. Carter SA, Tate RB. The value of toe pulse waves in determination of risks for limb amputation and death in patients with peripheral arterial disease and skin ulcers or gangrene. J Vasc Surg. 2001;33(4):708–14.

97. Cooper KJ, Peña C, Benenati J. Determining End Points for Critical Limb Ischemia Interventions. Tech Vasc Interv Radiol [Internet]. 2016;19(2):104–12. Available from: http://dx.doi.org/10.1053/j.tvir.2016.04.003

98. Ro DH, Moon HJ, Kim JH, Lee KM, Kim SJ, Lee DY. Photoplethysmography and continuous-wave Doppler ultrasound as a complementary test to ankle-brachial index in detection of stenotic peripheral arterial disease. Angiology. 2013 May;64(4):314–20.

99. Andrews KL, Dib MY, Shives TC, Hoskin TL, Liedl DA, Boon AJ. Noninvasive arterial studies including transcutaneous oxygen pressure measurements with the limbs elevated or dependent to predict healing after partial foot amputation. Am J Phys Med Rehabil. 2013;92(5):385–92.

149

100. Pardo M, Alcaraz M, Bernal FL, Felices JM, Achel GD, Breijo-Marquez FR, et al. A solution to ankle-brachial index limitations in peripheral transluminal angioplasty. Radiol Medica. 2013;118(8):1373–8.

101. Kalani. Transcutaneous Oxygen Tension and To e of Diabetic Foot Ulcers. 1999;22(1):1–5.

102. Apelqvist J, Bakker K, van Houtum WH, Schaper NC. Practical guidelines on the management and prevention of the diabetic foot: based upon the International Consensus on the Diabetic Foot (2007) Prepared by the International Working Group on the Diabetic Foot. Diabetes Metab Res Rev. 2008;24 Suppl 1:S181-7.

103. Kagaya Y, Ohura N, Suga H, Eto H, Takushima A, Harii K. “Real angiosome” assessment from peripheral tissue perfusion using tissue oxygen saturation foot-mapping in patients with critical limb ischemia. Eur J Vasc Endovasc Surg [Internet]. 2014;47(4):433–41. Available from: http://dx.doi.org/10.1016/j.ejvs.2013.11.011

104. Yudovsky D, Nouvong A, Schomacker K, Pilon L. Assessing diabetic foot ulcer development risk with hyperspectral tissue oximetry. J Biomed Opt [Internet]. 2011;16(2):026009. Available from: http://biomedicaloptics.spiedigitallibrary.org/article.aspx?doi=10.1117/1.3535592

105. Finlay JC, Foster TH. Hemoglobin oxygen saturations in phantoms and in vivo from measurements of steady-state diffuse reflectance at a single, short source-detector separation. Med Phys. 2004;31(7):1949–59.

106. Nouvong A, Hoogwerf B, Mohler E, Davis B, Tajaddini A, Medenilla E. Evaluation of diabetic foot ulcer healing with hyperspectral imaging of oxyhemoglobin and deoxyhemoglobin. Diabetes Care. 2009 Nov;32(11):2056–61.

107. Khaodhiar L, Dinh T, Schomacker KT, Panasyuk S V, Freeman JE, Lew R, et al. The Use of Medical Hyperspectral Technology to Evaluate Microcirculatory Changes in Diabetic Foot Ulcers and to Predict Clinical Outcomes. 2007;30(4):903–10. Available from: http://care.diabetesjournals.org/content/30/4/903

108. Gillies R, Freeman JE, Cancio LC, Brand D, Hopmeier M, Mansfield JR. Systemic Effects of Shock and Resuscitation Monitored by Visible Hyperspectral Imaging. Diabetes Technol Ther [Internet]. 2003;5(5):847–55. Available from: http://online.liebertpub.com/doi/abs/10.1089/152091503322527058

109. Yudovsky D, Nouvong A, Pilon L. Hyperspectral imaging in diabetic foot wound care. J Diabetes Sci Technol. 2010;4(5):1099–113.

110. Tuchin V. Tissue optics: light scattering methods and instruments for medical diagnosis, 2nd ed. San Diego (CA): SPIE Press. 2nd ed. 2007.

111. Zuzak KJ, Schaeberle MD, Lewis EN, Levin IW. Visible reflectance hyperspectral imaging: characterization of a noninvasive, in vivo system for determining tissue perfusion. Anal Chem. 2002 May;74(9):2021–8.

112. Rother U, Lang W. Noninvasive measurements of tissue perfusion in critical limb ischemia. Gefasschirurgie. 2018;23:8–12.

113. Harrison DK, McCollum PT, Newton DJ, Hickman P, Jain AS. Amputation level assessment using lightguide spectrophotometry. Prosthet Orthot Int. 1995 Dec;19(3):139–47.

114. Rother U, Kapust J, Lang W, Horch RE, Gefeller O, Meyer A. The Angiosome Concept Evaluated on the Basis of Microperfusion in Critical Limb Ischemia Patients-an Oxygen to See Guided Study. Microcirculation. 2015;22(8):737–43.

115. Rother U, Krenz K, Lang W, Horch RE, Schmid A, Heinz M, et al. Immediate changes of angiosome perfusion during tibial angioplasty. In: Journal of Vascular Surgery. 2017.

116. Montero-Baker MF, Au-Yeung KY, Wisniewski NA, Gamsey S, Morelli-Alvarez L, Mills JL, et al. The First-in-Man “si Se Puede” Study for the use of micro-oxygen sensors (MOXYs) to determine dynamic relative oxygen indices in the feet of

150

patients with limb-threatening ischemia during endovascular therapy. J Vasc Surg [Internet]. 2015;61(6):1501-1510.e1. Available from: http://dx.doi.org/10.1016/j.jvs.2014.12.060

117. Igari K, Kudo T, Toyofuku T, Jibiki M, Inoue Y, Kawano T. Quantitative evaluation of the outcomes of revascularization procedures for peripheral arterial disease using indocyanine green angiography. Eur J Vasc Endovasc Surg [Internet]. 2013;46(4):460–5. Available from: http://dx.doi.org/10.1016/j.ejvs.2013.07.016

118. Venermo M, Settembre N, Albäck A, Vikatmaa P, Aho PS, Lepäntalo M, et al. Pilot Assessment of the Repeatability of Indocyanine Green Fluorescence Imaging and Correlation with Traditional Foot Perfusion Assessments. Eur J Vasc Endovasc Surg [Internet]. 2016;52(4):527–33. Available from: http://dx.doi.org/10.1016/j.ejvs.2016.06.018

119. Cherrick GR, Stein SW, Leevy CM, Davidson CS. Indocyanine Green: Observations on Its Physical Properties, Plasma Decay, and Hepatic Extraction. J Clin Invest [Internet]. 1960;39(4):592–600. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC293343/

120. Settembre N, Kauhanen P, Albäck A, Spillerova K, Venermo M. Quality Control of the Foot Revascularization Using Indocyanine Green Fluorescence Imaging. World J Surg. 2017;41(7):1919–26.

121. Rother U, Lang W, Horch RE, Ludolph I, Meyer A, Regus S. Microcirculation Evaluated by Intraoperative Fluorescence Angiography after Tibial Bypass Surgery. Ann Vasc Surg [Internet]. 2017;40:190–7. Available from: http://dx.doi.org/10.1016/j.avsg.2016.07.084

122. Hodgkiss-Harlow KD, Bandyk DF. Interpretation of arterial duplex testing of lower-extremity arteries and interventions. Semin Vasc Surg [Internet]. 2013;26(2–3):95–104. Available from: http://dx.doi.org/10.1053/j.semvascsurg.2013.11.002

123. Collins R, Burch J, Cranny G, Aguiar-Ibáñez R, Craig D, Wright K, et al. Duplex ultrasonography, magnetic resonance angiography, and computed tomography angiography for diagnosis and assessment of symptomatic, lower limb peripheral arterial disease: Systematic review. Br Med J. 2007;334(7606):1257–61.

124. Mustapha JA, Saab F, Diaz-Sandoval L, Karenko B, McGoff T, Heaney C, et al. Comparison between angiographic and arterial duplex ultrasound assessment of tibial arteries in patients with peripheral arterial disease: on behalf of the Joint Endovascular and Non-Invasive Assessment of LImb Perfusion (JENALI) Group. J Invasive Cardiol. 2013 Nov;25(11):606–11.

125. Ntziachristos V, Razansky D. Molecular imaging by means of multispectral optoacoustic tomography (MSOT). Chem Rev. 2010 May;110(5):2783–94.

126. Plumb AA, Huynh NT, Guggenheim J, Zhang E, Beard P. Rapid volumetric photoacoustic tomographic imaging with a Fabry-Perot ultrasound sensor depicts peripheral arteries and microvascular vasomotor responses to thermal stimuli. Eur Radiol. 2018;28(3):1037–45.

127. Fotiadis N, Kyriakides C, Bent C, Vorvolakos T, Matson M. 64-section CT angiography in patients with critical limb ischaemia and severe claudication: Comparison with digital subtractive angiography. Clin Radiol [Internet]. 2011;66(10):945–52. Available from: http://dx.doi.org/10.1016/j.crad.2011.04.015

128. Huang SY, Nelson RC, Miller MJ, Kim CY, Lawson JH, Husarik DB, et al. Assessment of Vascular Contrast and Depiction of Stenoses in Abdominopelvic and Lower Extremity Vasculature. Comparison of Dual-energy MDCT with Digital Subtraction Angiography. Acad Radiol [Internet]. 2012;19(9):1149–57. Available from: http://dx.doi.org/10.1016/j.acra.2012.04.014

129. Hansmann J, Morelli JN, Michaely HJ, Riester T, Budjan J, Schoenberg SO, et al. Nonenhanced ECG-gated quiescent-interval single shot MRA: Image quality and

151

stenosis assessment at 3 tesla compared with contrast-enhanced MRA and digital subtraction angiography. J Magn Reson Imaging. 2014;39(6):1486–93.

130. Kinlay S. Management of critical limb ischemia. Circ Cardiovasc Interv. 2016;9(2):1–11.

131. Grözinger G, Pohmann R, Schick F, Grosse U, Syha R, Brechtel K, et al. Perfusion measurements of the calf in patients with peripheral arterial occlusive disease before and after percutaneous transluminal angioplasty using mr arterial spin labeling. J Magn Reson Imaging. 2013;40(4):980–7.

132. Pollak AW, Meyer CH, Epstein FH, Jiji RS, Hunter JR, Dimaria JM, et al. Arterial spin labeling mr imaging reproducibly measures peak-exercise calf muscle perfusion: A study in patients with peripheral arterial disease and healthy volunteers. JACC Cardiovasc Imaging [Internet]. 2012;5(12):1224–30. Available from: http://dx.doi.org/10.1016/j.jcmg.2012.03.022

133. Kinlay S. Management of critical limb ischemia. Circ Cardiovasc Interv. 2016;9(2). 134. Marston WA, Davies SW, Armstrong B, Farber MA, Mendes RC, Fulton JJ, et al.

Natural history of limbs with arterial insufficiency and chronic ulceration treated without revascularization. J Vasc Surg. 2006;44(1):108–15.

135. Elgzyri T, Larsson J, Thörne J, Eriksson KF, Apelqvist J. Outcome of ischemic foot ulcer in diabetic patients who had no invasive vascular intervention. Eur J Vasc Endovasc Surg [Internet]. 2013;46(1):110–7. Available from: http://dx.doi.org/10.1016/j.ejvs.2013.04.013

136. Teraa M, Conte MS, Moll FL, Verhaar MC. Critical limb ischemia: Current trends and future directions. J Am Heart Assoc. 2016;5(2):1–9.

137. Goodney PP, Beck AW, Nagle J, Welch HG, Zwolak RM. National trends in lower extremity bypass surgery, endovascular interventions, and major amputations. J Vasc Surg. 2009 Jul;50(1):54–60.

138. Jones WS, Patel MR, Dai D, Subherwal S, Stafford J, Calhoun S, et al. Temporal trends and geographic variation of lower-extremity amputation in patients with peripheral artery disease: results from U.S. Medicare 2000-2008. J Am Coll Cardiol. 2012 Nov;60(21):2230–6.

139. Egorova NN, Guillerme S, Gelijns A, Morrissey N, Dayal R, McKinsey JF, et al. An analysis of the outcomes of a decade of experience with lower extremity revascularization including limb salvage, lengths of stay, and safety. J Vasc Surg. 2010 Apr;51(4):878–85, 885.e1.

140. Benoit E, O’Donnell TFJ, Kitsios GD, Iafrati MD. Improved amputation-free survival in unreconstructable critical limb ischemia and its implications for clinical trial design and quality measurement. J Vasc Surg. 2012 Mar;55(3):781–9.

141. Hong MS, Beck AW, Nelson PR. Emerging national trends in the management and outcomes of lower extremity peripheral arterial disease. Ann Vasc Surg. 2011 Jan;25(1):44–54.

142. Aboyans V, Ricco J-B, Bartelink M-LEL, Björck M, Brodmann M, Cohnert T, et al. 2017 ESC Guidelines on the Diagnosis and Treatment of Peripheral Arterial Diseases, in collaboration with the European Society for Vascular Surgery (ESVS). Eur Heart J [Internet]. 2017;(June):763–821. Available from: http://academic.oup.com/eurheartj/article/doi/10.1093/eurheartj/ehx095/4095038/2017-ESC-Guidelines-on-the-Diagnosis-and-Treatment

143. Slovut DP, Lipsitz EC. Surgical technique and peripheral artery disease. Circulation. 2012 Aug;126(9):1127–38.

144. Pomposelli FB, Kansal N, Hamdan AD, Belfield A, Sheahan M, Campbell DR, et al. A decade of experience with dorsalis pedis artery bypass: analysis of outcome in more than 1000 cases. J Vasc Surg. 2003 Feb;37(2):307–15.

145. Ambler GK, Twine CP. Graft type for femoro-popliteal bypass surgery. Cochrane

152

database Syst Rev. 2018 Feb;2:CD001487. 146. Conte MS. Critical appraisal of surgical revascularization for critical limb ischemia.

In: Journal of Vascular Surgery [Internet]. Society for Vascular Surgery; 2013. p. 8S-13S. Available from: http://dx.doi.org/10.1016/j.jvs.2012.05.114

147. Pereira CE, Albers M, Romiti M, Brochado-Neto FC, Pereira CAB. Meta-analysis of femoropopliteal bypass grafts for lower extremity arterial insufficiency. J Vasc Surg. 2006 Sep;44(3):510–7.

148. Schanzer A, Hevelone N, Owens CD, Belkin M, Bandyk DF, Clowes AW, et al. Technical factors affecting autogenous vein graft failure: observations from a large multicenter trial. J Vasc Surg. 2007 Dec;46(6):1180–90; discussion 1190.

149. Tips and tricks for a correct “endo approach.” J Cardiovasc SURG. 2013;54(6):685–711.

150. Silvestro M, Palena LM, Manzi M, Gómez-Jabalera E, Vishwanath D, Casini A, et al. Anterolateral retrograde access to the distal popliteal artery and to the tibioperoneal trunk for recanalization of femoropopliteal chronic total occlusions. Journal of Vascular Surgery [Internet]. 2018;1–9. Available from: https://doi.org/10.1016/j.jvs.2018.05.231

151. Palena LM, Manzi M. Extreme Below-the-Knee Interventions: Retrograde Transmetatarsal or Transplantar Arch Access for Foot Salvage in Challenging Cases of Critical Limb Ischemia. J Endovasc Ther [Internet]. 2012;19(6):805–11. Available from: http://jet.sagepub.com/lookup/doi/10.1583/JEVT-12-3998R.1

152. Lumsden a B, Davies MG, Peden EK. Medical and endovascular management of critical limb ischemia. J Endovasc Ther [Internet]. 2009;16(2 Suppl 2):II31-62. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19624074

153. Cassese S, Byrne RA, Ott I, Ndrepepa G, Nerad M, Kastrati A, et al. Paclitaxel-coated versus uncoated balloon angioplasty reduces target lesion revascularization in patients with femoropopliteal arterial disease: a meta-analysis of randomized trials. Circ Cardiovasc Interv. 2012 Aug;5(4):582–9.

154. Dake MD, Ansel GM, Jaff MR, Ohki T, Saxon RR, Smouse HB, et al. Paclitaxel-eluting stents show superiority to balloon angioplasty and bare metal stents in femoropopliteal disease: twelve-month Zilver PTX randomized study results. Circ Cardiovasc Interv. 2011 Oct;4(5):495–504.

155. Tepe G, Laird J, Schneider P, Brodmann M, Krishnan P, Micari A, et al. Drug-coated balloon versus standard percutaneous transluminal angioplasty for the treatment of superficial femoral and popliteal peripheral artery disease: 12-month results from the IN.PACT SFA randomized trial. Circulation. 2015 Feb;131(5):495–502.

156. Rosenfield K, Jaff MR, White CJ, Rocha-Singh K, Mena-Hurtado C, Metzger DC, et al. Trial of a Paclitaxel-Coated Balloon for Femoropopliteal Artery Disease. N Engl J Med. 2015 Jul;373(2):145–53.

157. Katsanos K, Spiliopoulos S, Kitrou P, Krokidis M, Karnabatidis D. Risk of Death Following Application of Paclitaxel-Coated Balloons and Stents in the Femoropopliteal Artery of the Leg: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. J Am Heart Assoc. 2018 Dec;7(24):e011245.

158. Bradbury AW, Adam DJ, Beard JD, Cleveland T, Forbes JF, Fowkes FGR, et al. Bypass versus angioplasty in severe ischaemia of the leg (BASIL): Multicentre, randomised controlled trial. Lancet. 2005;366(9501):1925–34.

159. Conte MS, Geraghty PJ, Bradbury AW, Hevelone ND, Lipsitz SR, Moneta GL, et al. Suggested objective performance goals and clinical trial design for evaluating catheter-based treatment of critical limb ischemia. J Vasc Surg [Internet]. 2009;50(6):1462-1473.e3. Available from: http://dx.doi.org/10.1016/j.jvs.2009.09.044

160. Patel MR, Conte MS, Cutlip DE, Dib N, Geraghty P, Gray W, et al. Evaluation and

153

treatment of patients with lower extremity peripheral artery disease: Consensus definitions from peripheral academic research consortium (PARC). J Am Coll Cardiol. 2015;65(9):931–41.

161. Menard MT, Farber A, Assmann SF, Choudhry NK, Conte MS, Creager MA, et al. Design and Rationale of the Best Endovascular Versus Best Surgical Therapy for Patients With Critical Limb Ischemia (BEST-CLI) Trial. J Am Heart Assoc. 2016;5(7):1–21.

162. Farber A. An Update on the BEST-CLI Trial [Internet]. Vol. 14, Vascular Disease Management. 2017. p. 166–7. Available from: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=9063184

163. Dormandy J, Heeck L, Vig S. Major amputations: clinical patterns and predictors. Semin Vasc Surg. 1999 Jun;12(2):154–61.

164. Mudge E, Price P, Neal W, Harding KG. A randomized controlled trial of larval therapy for the debridement of leg ulcers: Results of a multicenter, randomized, controlled, open, observer blind, parallel group study. Wound Repair Regen. 2014;22(1):43–51.

165. Conte MS. Understanding objective performance goals for critical limb ischemia trials. Semin Vasc Surg [Internet]. 2010;23(3):129–37. Available from: http://dx.doi.org/10.1053/j.semvascsurg.2010.06.001

166. Palena LM, Garcia LF, Brigato C, Sultato E, Candeo A, Baccaglini T, et al. Angiosomes: how do they affect my treatment? Tech Vasc Interv Radiol [Internet]. 2014 Sep 1 [cited 2018 Mar 9];17(3):155–69. Available from: http://www.ncbi.nlm.nih.gov/pubmed/25241316

167. Taylor GI, Palmer JH. The vascular territories (angiosomes) of the body: experimental study and clinical applications. Br J Plast Surg. 1987;40(2):113–41.

168. Alexandrescu V-A, Hubermont G, Philips Y, Guillaumie B, Ngongang C, Vandenbossche P, et al. Selective primary angioplasty following an angiosome model of reperfusion in the treatment of Wagner 1-4 diabetic foot lesions: practice in a multidisciplinary diabetic limb service. J Endovasc Ther [Internet]. 2008;15(5):580–93. Available from: http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=med5&NEWS=N&AN=18840046%5Cnhttp://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=emed8&NEWS=N&AN=2008545157

169. Alexandrescu V, Hubermont G. The challenging topic of diabetic foot revascularization: Does the angiosome-guided angioplasty may improve outcome. J Cardiovasc Surg (Torino). 2012;Feb;53(1):3-12. Review.

170. Serra R, Grande R, Scarcello E, Buffone G, de Franciscis S. Angiosome-targeted revascularisation in diabetic foot ulcers. Int Wound J. 2015;12(5):555–8.

171. Chae KJ, Shin JY. Is angiosome-targeted angioplasty effective for limb salvage and wound healing in diabetic foot?: A meta-analysis. PLoS One. 2016;11(7):1–11.

172. Lejay A, Georg Y, Tartaglia E, Gaertner S, Geny B, Thaveau F, et al. Long-term outcomes of direct and indirect below-the-knee open revascularization based on the angiosome concept in diabetic patients with critical limb ischemia. Ann Vasc Surg. 2014 May;28(4):983–9.

173. Söderström M, Albäck A, Biancari F, Lappalainen K, Lepäntalo M, Venermo M. Angiosome-targeted infrapopliteal endovascular revascularization for treatment of diabetic foot ulcers. J Vasc Surg [Internet]. 2013;57(2):427–35. Available from: http://dx.doi.org/10.1016/j.jvs.2012.07.057

174. Alexandrescu V, Hubermont G. Primary infragenicular angioplasty for diabetic neuroischemic foot ulcers following the angiosome distribution: a new paradigm for the vascular interventionist? Diabetes Metab Syndr Obes. 2011;4:327–36.

154

175. Alexandrescu V, Vincent G, Azdad K, Hubermont G, Ledent G, Ngongang C, et al. A Reliable Approach to Diabetic Neuroischemic Foot Wounds: Below-the-Knee Angiosome-Oriented Angioplasty. J Endovasc Ther [Internet]. 2011;18(3):376–87. Available from: http://jet.sagepub.com/lookup/doi/10.1583/10-3260.1

176. Bosanquet DC, Glasbey JCD, Williams IM, Twine CP. Systematic review and meta-analysis of direct versus indirect angiosomal revascularisation of infrapopliteal arteries. Eur J Vasc Endovasc Surg [Internet]. 2014;48(1):88–97. Available from: http://dx.doi.org/10.1016/j.ejvs.2014.04.002

177. Yamada T, Gloviczki P, Bower TC, Naessens JM, Carmichael SW. Variations of the arterial anatomy of the foot. Am J Surg. 1993 Aug;166(2):130–5; discussion 135.

178. Kropman RHJ, Kiela G, Moll FL, de Vries J-PPM. Variations in anatomy of the popliteal artery and its side branches. Vasc Endovascular Surg. 2011 Aug;45(6):536–40.

179. Zheng M, Chen C, Qiu Q, Wu C. Ultrasound in diagnosis of anatomical variation of anterior and posterior tibial arteries. Med Ultrason. 2016 Mar;18(1):64–9.

180. Utsunomiya M, Nakamura M, Nakanishi M, Takagi T, Hara H, Onishi K, et al. Impact of wound blush as an angiographic end point of endovascular therapy for patients with critical limb ischemia. J Vasc Surg [Internet]. 2012;55(1):113–21. Available from: http://dx.doi.org/10.1016/j.jvs.2011.08.001

181. Jens S, Marquering HA, Koelemay MJW, Reekers JA. Perfusion Angiography of the Foot in Patients with Critical Limb Ischemia: Description of the Technique. Cardiovasc Intervent Radiol. 2015;38(1):201–5.

182. Philips. Tips & tricks for instant 2D Perfusion images in Peripheral Artery Disease. 2016;Oct(4522 991 22561):1–8.

183. Reekers JA, Koelemay MJW, Marquering HA, van Bavel ET. Functional Imaging of the Foot with Perfusion Angiography in Critical Limb Ischemia. Cardiovasc Intervent Radiol. 2016;39(2):183–9.

184. Manzi M, Cester G, Palena LM, Alek J, Candeo A, Ferraresi R. Vascular Imaging of the Foot: The First Step toward Endovascular Recanalization. RadioGraphics [Internet]. 2011;31(6):1623–36. Available from: http://pubs.rsna.org/doi/10.1148/rg.316115511

185. Toursarkissian B, D’Ayala M, Stefanidis D, Shireman PK, Harrison A, Schoolfield J, et al. Angiographic scoring of vascular occlusive disease in the diabetic foot: Relevance to bypass graft patency and limb salvage. J Vasc Surg. 2002;35(3):494–500.

186. Reed GW, Salehi N, Giglou PR, Kafa R, Malik U, Maier M, et al. Time to Wound Healing and Major Adverse Limb Events in Patients with Critical Limb Ischemia Treated with Endovascular Revascularization. Ann Vasc Surg [Internet]. 2016;36:190–8. Available from: http://dx.doi.org/10.1016/j.avsg.2016.02.032

187. Hinrichs JB, Murray T, Akin M, Lee M, Brehm MU, Wilhelmi M, et al. Evaluation of a novel 2D perfusion angiography technique independent of pump injections for assessment of interventional treatment of peripheral vascular disease. Int J Cardiovasc Imaging. 2017;33(3):295–301.

188. Kostrzewa M, Kara K, Pilz L, Mueller-Muertz H, Rathmann N, Schoenberg SO, et al. Treatment Evaluation of Flow-Limiting Stenoses of the Superficial Femoral and Popliteal Artery by Parametric Color-Coding Analysis of Digital Subtraction Angiography Series. Cardiovasc Intervent Radiol. 2017;40(8):1147–54.

189. Murray T, Rodt T, Lee MJ. Two-dimensional perfusion angiography of the foot: Technical considerations and initial analysis. J Endovasc Ther. 2016;23(1):58–64.

190. Hill NR, Fatoba ST, Oke JL, Hirst JA, O’Callaghan CA, Lasserson DS, et al. Global Prevalence of Chronic Kidney Disease - A Systematic Review and Meta-Analysis. PLoS One. 2016;11(7):e0158765.

155

191. Friedman HZ, DeBoe SF, McGillem MJ, Mancini GB. Immediate effects of graded ionic and nonionic contrast injections on coronary blood flow and myocardial function. Implications for digital coronary angiography. Invest Radiol. 1987 Sep;22(9):722–7.

192. Bollinger A, Breddin K, Hess H, Heystraten FMJ, Kollath J, Konttila A, et al. Semiquantitative assessment of lower limb atherosclerosis from routine angiographic images. Atherosclerosis. 1981;38(3–4):339–46.

193. Graziani L, Silvestro A, Bertone V, Manara E, Andreini R, Sigala A, et al. Vascular Involvement in Diabetic Subjects with Ischemic Foot Ulcer: A New Morphologic Categorization of Disease Severity. Eur J Vasc Endovasc Surg. 2007;33(4):453–60.

194. Bargellini I, Piaggesi A, Cicorelli A, Rizzo L, Cervelli R, Iacopi E, et al. Predictive value of angiographic scores for the integrated management of the ischemic diabetic foot. J Vasc Surg [Internet]. 2013;57(5):1204–12. Available from: http://www.embase.com/search/results?subaction=viewrecord&from=export&id=L52396918%0Ahttp://dx.doi.org/10.1016/j.jvs.2012.10.104

195. Bellmunt-montoya S. The Problem of Multiple Hypothesis Testing and the Assumption of Causality. Eur J Vasc Endovasc Surg [Internet]. 2018;8035. Available from: https://doi.org/10.1016/j.ejvs.2018.09.024

157

9. LIST OF FIGURES

159

Page Figure 1. Arterial wall structure. 43

Figure 2. Atherothrombotic disease progression. 43

Figure 3. Cardiovascular disease prevalence and presentation. 44

Figure 4. Diagram of mechanisms leading to vascular calcification. 47

Figure 5. Diagram of X-ray 47

Figure 6. Aortoiliac TASC classification 52

Figure 7. Femoropopliteal TASC classification 52

Figure 8. Below-the-knee TASC classification 53

Figure 9. WIfI consensus classification 55

Figure 10. Probability of ulcer healing 59

Figure 11. Hyperspectral imaging image sample 60

Figure 12. O2C measuring principle 61

Figure 13. The O2C probe 62

Figure 14. Fluorescence angiography image 63

Figure 15. Time-intensity curves from ICG-FI 63

Figure 16. CLI treatment algorithm 68

Figure 17. Step-by-step approach to chronic total occlusions 69

Figure 18. Angiosome skin-surface representation 71

Figure 19. Color mapping of perfusion angiography 73

Figure 20. Representation of the PA parameters over the curve 75

Figure 21. Formula for the global limb ATS. SFA 95

Figure 22. Examples of the Abano Terme scoring system 96

Figure 23. Algorithm of exclusions 101

Figure 24. Subgroups for analysis of the PA 102

Figure 25. Ulcer healing rate 104

Figure 26. TASC classifications of the before and after treatment angiographies 105

Figure 27. Evolution of WIfI clinical stage 105

Figure 28. Box-plots of ATSs 106

160

Page

Figure 29a. Box-plots of PA parameters 107

Figure 29b. Box-plots of PA parameters 107

Figure 30. Receiver Operating Curves for the post-EVT PA parameters 111

Figure 31. Kaplan-Meier survival curves for TTH 113

Figure 32. Multivariate ROC curve 114

Figure 33. Histograms of TcPO2 measures 116

Figure 34. Plastination of microcirculation in the foot 126

Figure 35. Decision algorithm to apply PA 131

161

10. LIST OF TABLES

163

Page

Table 1. Rutherford classification 49

Table 2. University of Texas Ulcer Classification 50

Table 3. Baseline characteristics of cases for analysis 103

Table 4. Laboratory tests 103

Table 5. EVT variables 103

Table 6. Baseline clinical stages frequencies. 103

Table 7. Changes in TASC, ATS and PA parameters 106

Table 8. Distribution of ∆ATS and ∆PA parameters 108

Table 9. Follow-up events frequencies 108

Table 10. Baseline characteristics of the subgroup of patients without ulcers 109

Table 11. Differences in post-EVT and ∆PA parameters for TLR during follow-up 109

Table 12. Comorbidities and possible confounding factors among cases without ulcers 109

Table 13. Comorbidities and possible confounding factors 110

Table 14. Differences in post-EVT and ∆PA parameters for TTH at 30 days 110

Table 15. Chi Square analysis for cut-off points 112



Table 18. Multivariate analysis with possible confounding factors 114

Table 19. ATS predicting power for wound healing 115

Table 20. Spearman’s rho correlation between PA parameters and classification systems 115

Table 21. Relation between the PA parameters and the state of the plantar arch 117

Table 22. Compared diagnostic methods 130

165

11. ANNEXES

167

11.1. Examples of perfusion angiography images

PA 1. Clearly good results. In every case of these ones the result obtained was optimal, both considering the color-coded image and the time-density plot, but this alone would still be a subjective impression nor could be stated the exact amount of improvement ne-cessary to obtain satisfying clinical outcomes.

PA 2. Artifacts. Movement artifacts could be identified in these pictures by a high peak-valley (A), movement in the bone (B) or solely in the toes (C), that can be corrected ex-cluding the artifact from the ROI. In other cases, artifacts could not be seen in the image but still be affecting to the curve’s shape (D), which cannot be corrected either since it could be affecting parameter values.

168

PA 3. Effect of physician-controlled length of exposure, suppose that the area under the curve could not be compared. These examples show an exposure length differences of 5.6 s (A), 5 (B), 4.7 s (C) and -9.3 (D), which usually uses a longer exposure before the treatment because a delayed arrival of the contrast media.

PA 4. Perfusion in wound-blush. When a wound-blush effect, a typically sing of good prognosis, the time-density curve adopts usually a high-plateau shape (A and B). However, it cannot be triggered and it is not constant (C and D), being better detected in the color-shape image and without any significative difference from conventional angio-graphy.

169

PA 5. Vasospasm. When an after treatment a distal vasospasm occurred, a diminishment of the maximum density is observed; either with a faster arrival of the contrast (A and B) or just stopping the ascent of the curve (C and D).

PA 6. Patient-dependent shape of the curve is shown in these samples, as a parallel as-cending curves (A and D) or similar shape of the whole plateau (C and B). It can also be seen in the same patient different ROIs (F and G), where both initial and completion runs have similar shapes.

170

11.2. Presented abstract: Hemodynamic significance of perfusion

angiography parameters.

Gómez, E1, 2; Mestres, G1; Palena, LM2; Manzi, M2 1Hospital Clínic de Barcelona. 2Policlinico Abano Terme

Introduction: Perfusion angiography (PA) is a post-processing software algorithm that does not require digital subtraction angiography (DSA) or contrast medium injection. Given that their significance is unknown, PA parameters are not being used in decision-making pro-cesses. Objectives: Our study is a first attempt to demonstrate that PA parameters may provide val-uable hemodynamic significance. Material used: On consecutive patients undergoing endovascular treatment (EVT) at Poli-clinico Abano Terme (Italy) for critical limb ischemia (CLI), standard angiographic studies of the limb and PA of the foot before undergoing EVT and thereafter were performed, meas-uring all its parameters (arrival time, time to peak, wash-in rate, width, area under the curve, and mean transit time). Methodology: Demographic data and PA analysis on the foot were measured and TASC classification in femoro-popliteal (FP-TASC) and below the knee (BTK-TASC) segments were assigned accordingly. Results: 74 consecutive patients were studied. Mean age was 71, 74% were men. 6 patients were excluded due to PA artefacts. All PA parameters showed significant improvements be-tween PA performed previously and after EVT (p<0.03), according to angiographic find-ings. Wash-in rate was inversely related with both FP-TASC (p=0.026) and BTK-TASC (p=0.009) classifications, and arrival time had a direct relation with BTK-TASC (p=0.032). The differences in BTK-TASC after EVT revealed a negative correlation with arrival time (p=0.005) and width (p=0.045). Conclusions: Arrival time appears to be the most related PA parameter with BTK-TASC (for an isolated angiography and for differences after EVT). Wash-in rate seems to be more representative of the whole TASC classification of the limb, both FP-TASC and BTK-TASC.

Presented at the SITE 2017 and SCACVE congress 2017

171

11.3. Presented abstract: Perfusion angiography in the prediction of wound healing in endovascular treatment of critical limb ischemia. Gómez Jabalera, Efrem1,2; Mestres, Gaspar2; Bellmunt, Sergi3; Palena, L Mariano1; Manzi, Marco11

Policlinico Abano Terme; 2Hospital Clínic de Barcelona; 3Hospital Universitari Vall d'Hebron

Introduction: Current endovascular treatments (EVT) of critical limb ischemia (CLI) have no clear treatment goals. Perfusion Angiography (PA) is an image-processing software that analyzes density per pixel over time, and it could predict the clinical success of EVT.

Methods: Consecutive patients undergoing EVT for CLI were analyzed with PA before and after treatment. Exclusion criteria were poor PA image quality, no ulcer, death and loss at follow-up. Demographic and clinical data were recorded and clinical follow-up was performed to trace the time to heal (TTH) of the ulcers. The PA parameters were measured on the contrast time-density curve: Arrival Time (AT), Peak Time (PT), Wash-in Rate, Width, Area Under Curve and Mean Transit Time (MTT). Two cohorts based upon a TTH of less (group A) or more than 30 days (group B). We used Student-t test for independent variables and for changes before and after EVT and cut-off values from ROC curves were identified.

Results: From January2015 to July2016, 332 consecutive patients were studied and 123 were excluded. Mean age was 72 years and 67.5% were men. 133 patients had Rutherford 5 and 76 had Rutherford 6 lesions, with similar distribution in both groups. We found significant differences between groups in the following after treatment PA parameters: AT (+1.2seconds, 95%CI:2.01-0.39; p = .004), PT (+0.56seconds, 95% CI: 1.01-0.11; p = 0.014) and MTT (+0.5seconds, 95%CI: 1-0.08; p = .022), and also on the improvement of MTT (+0.64 seconds, 95% CI: 1.16-0.1; p = .02). Cut-off values were: AT > 6 seconds (Sensitivity = 64.8%, Specificity = 59%, p = .001), PT > 5.2 seconds (Sensitivity = 43.7%, Specificity = 76.9%, p = .002), and improvement of MTT > 1.7 seconds (Sensitivity = 42.4%, Specificity = 81.7%, p = .016).

Conclusion: AT, PT and MTT can predict CLI wound healing in less than 30 days.

Presented at SCACVE 2018 and the 32nd ESVS congress 2018

172

11.4. Presented abstract: Angiography based Abano Terme Score: a novel descriptive system for below the groin arterial disease.

Efrem Gómez-Jabalera1,2, Sergi Bellmunt Montoya3, Jaume Dilmé Muñoz4, Vicenç Artigas Raventós5, Luis Mariano Palena2, Marco Manzi2

1Hospital Clínic Barcelona, Vascular Surgery, Barcelona, Spain, 2Policlinico Abano Terme, Interventional radiology, Abano Terme, Italy, 3Hospital General Universitari Vall d'Hebron, Vascular surgery, Barcelona, Spain, 4Hospital de la Santa creu i Sant Pau, Vascular Surgery, Barcelona, Spain, 5Universitat Autònoma de Barcelona, Surgery, Barcelona, Spain

Introduction: The TASC classification is the most frequently used system to describe lower limb angiographic lesions. However, its lack in hemodynamic and clinical information render it a suboptimal system for daily practice. The calf and foot score of the Joint Vascular Society Council (JVSC) provides a grading system of bypass grafting outflow and foot steno-occlusions. Inspired on the latter score, our aim is to describe the novel classification system Abano Terme Score (ATS), which adequately assesses the anatomy and burden of steno-occlusions in the whole limb, to ultimately select the best therapy for each patient.

Methods: The JVSC classification2 (0:no stenosis>20%; 1:21%-49% stenosis; 2:50%-99% stenosis; 2.5:<half the vessel length occluded; 3:>half the vessel length occluded) was adapted to each limb’s arteries: superficial femoral artery (SFA), popliteal, anterior tibial (ATA), posterior tibial (PTA) and peroneal arteries (PA). The limb-ATS was obtained by adding the above the knee (ATK) and the below the knee (BTK) scores. The ATK-ATS was calculated by adding the squares of the scores of SFA and popliteal artery. The BTK-ATS was calculated by multiplying the scores of ATA and PTA, adding the score for PA afterwards. Retrospective cohort analysis was done over endovascular treatment of critical limb ischemia patients which were studied with perfusion angiography; their baseline co-morbidities recorded, and pre-treatment and completion angiographies were performed and classified according to TASC (ATK and BTK) and ATS (ATK-ATS, BTK-ATS and whole limb-ATS). Perfusion angiography (Philips 2D Perfusion 1.1.6.) runs were performed in both angiographies and the arrival times (AT) were measured. Changes throughout treatment, perfusion angiography and TcPO2 at baseline and 1-month post-intervention were recorded. We assessed statistical post-revascularization changes and explored correlations (Spearman’s rho, sr) between TASC, ATS, AT and TcPO2.

Results: From January 2015 to July 2016, 246 patients received endovascular treatment and were studied with perfusion angiography (68.6% men, mean age of 72.2±10.4 years). In this cohort 92.8% were diabetics, 50.2% had chronic kidney disease and 9.2% end-stage renal disease. At 6 months, ulcer healing rate was 70.6%, amputation free-survival was 96.6%, and mortality of 4.2%. Student t-test showed statistical differences (p< 0.001) between ATS values before and after treatment in ATK region (from 4.7±4.6 to 0.3±0.7), BTK region (from 6±3.9 to 1.09±2.1) and on the whole limb (from 10.6±5.8 and to 1.43±2.3). The TcPO2 at baseline and at 1-month were 22±15.8mmHg, 50.9±11.9mmHg, respectively; with significant correlations between TcPO2 and the BTK-ATS (sr=0.25, p=0.020) and with whole limb-ATS (sr=0.24, p=0.023). Comparing both pre-treatment and completion perfusion angiographies, we found correlations between the AT and the ATK-TASC (sr=0.28, p< 0.001), the BTK-TASC (sr=0.38, p< 0.001) and the whole limb-ATS (sr=0.41, p< 0.001).

Conclusion: ATS is a simpler, easier and more thorough classification method than current systems available to describe the steno-occlusive burden of a target limb. It varies with revascularization, and appears to be mild-moderately correlated with TcPO2 and perfusion angiography (better than TASC in the latter). Further studies are warranted to validate and compare the ATS with other anatomical and hemodynamic diagnostic methods like doppler ultrasound and other patient-centered outcomes.

Presented at the 33rd ESVS congress

173

11.5. Article submitted to the European Journal of Vascular & Endovascular Surgery:

Clinical application of 2D perfusion angiography in critical ...

Documents

Transcript of Clinical application of 2D perfusion angiography in critical ...